Anthropic said its AI systems could drastically reshape the software and cybersecurity sectors in the “near term,” while also laying out a litany of cybersecurity and regulatory risks facing the company as it prepares to go public.

In a filing with the Securities and Exchange Commission ahead of its initial public offering, the company said its technology could transform entire industries and the economy. Already, cybersecurity-focused models by Anthropic and rival OpenAI can find and exploit cybersecurity vulnerabilities at an alarming speed.

But the company also acknowledged serious risks its models, including Claude, pose to itself, its users and the world. Hacks by bad actors using Claude or AI agents could cause financial, legal or physical damage, spark lawsuits and affect Anthropic’s ability to insure against future losses. 

Anthropic expects bad actors, including foreign governments, to increase their efforts to access its latest frontier models as their cybersecurity capabilities improve. That creates persistent risk, since its models could be used in cyberattacks carried out by foreign governments and cybercriminal groups.

“Any successful attempt to misuse, or any perceived misuse of, our models, platform, and applications for a cyberattack or for any other malicious purpose could result in exposure to liability, disruption of our customer or other commercial relationships, or damage to our reputation,” Anthropic said in its filing.

That potential legal and regulatory risk extends to independent security researchers and others who have “used, and may in the future use, our models to identify and exploit vulnerabilities in the systems of other companies, including other AI developers, at times in connection with authorized bug bounty programs.”

Anthropic cited the rise of AI-enabled cybersecurity incidents in the U.S., the U.K. and other countries, saying they could result in more adversarial regulatory environments around the world. European laws, for example, require detailed breach reports within a strict time frame, which could be difficult for Anthropic to meet.

Federal intervention could also disrupt AI development and release timelines. Anthropic cited the Department of Commerce placing export controls on its Mythos 5 and Fable 5 models, and confirmed that the Federal Trade Commission issued it a civil investigative demand earlier this month for more information on the safety and security of its products, as well as its consumer, data and privacy protection policies.

AI development will be guided by more than ethics

Anthropic has developed a public reputation as an ethics-first AI company,  most notably in disputes with the Department of Defense over how its models are used in wartime. But the IPO filing states the company must “balance safety considerations against commercial objectives, competitive pressures, and stakeholder expectations.”

The company also said it is “actively” improving security infrastructure and controls. It also has acknowledged it has moved forward on projects with “known limitations or gaps in our defenses” and may do so again.

“There can be no assurance, however, that the safeguards or access regimes we adopt will prove adequate, that we will correctly identify which capabilities require restriction, or that constraints we impose will not be circumvented,” the document stated.

AI agents can create new and “distinct” cybersecurity attack surfaces because they interact with so many other tools, databases, APIs and other third-party systems. That creates  feedback loops between a model and other systems, which can be compromised or manipulated to feed malicious instructions to the agent.

Anthropic suggested the problems could be so pronounced that it may have difficulty obtaining insurance for some cybersecurity incidents and data breaches related to its models.  

Andrew Yoon, head of research at the nonprofit CivAI, told CyberScoop that both the document and Anthropic’s overall record show a company that has consistently put profit considerations over building safer AI.

“I think in general they’re not meeting the moment here,” Yoon said. “Despite all of these risks that they flagged, they’re moving ahead pretty much as fast as humanly possible, and their justification for this is ‘You can’t do safety from second place.’”

Yoon said Anthropic has not only pushed its own frontier research toward dangerous and harmful capabilities like hacking, “they are pushing very aggressively ahead” and in some cases influencing competitors to follow suit. Anthropic’s release of its Mythos and Fable models earlier this year caused OpenAI to shift “its entire corporate strategy” to match the capabilities of Anthropic models in coding, science, and general intelligence by building models at breakneck speed.

“They have pretty directly caused the industry to speed up toward this mad, frenzied race and yet they’re saying ‘Oh, our models are too powerful,’ ”….it just doesn’t add up to me,” said Yoon.

But Yoon said it’s not clear that building AI systems to carry out dangerous, complex activities like hacking or bioweapons is necessary to be competitive in the global marketplace. He pointed to the wide assortment of Chinese-developed, open-weight AI models that are widely used around the world, but are not powerful enough to chain together complex exploits or develop new viruses.  

“This is not an accident, this is not the models generalizing from regular programming capabilities…the labs are intentionally training them on hacking as well,” said Yoon. “This is because they want to offer better defensive capabilities, they want to be able to sell offensive products to the government…and they do pretty cavalier testing with [the models] inside the labs.”

The post All the security risks Anthropic is worried about before its IPO  appeared first on CyberScoop.

A new bipartisan House bill would charge the federal government with developing formal guidance for defending AI models against distillation attacks.

The AI Cyber Defense Act, shared exclusively with CyberScoop ahead of release, is sponsored by Reps. Pat Harrigan, R-N.C., and George Whitesides, D-Calif. It would direct the National Institute for Standards and Technology to develop voluntary industry standards and best practices for preventing distillation attacks from foreign entities from four countries: Russia, China, Iran and North Korea.

In an interview, Harrigan told CyberScoop it is “incredibly expensive” for the U.S. AI industry to maintain its six-to-12 month lead over other countries, and distillation attacks give other parties an incentive to copy other models while keeping their own spending low.

He said he believes countries such as China are making a strategic choice to hold back on full investment of their own AI industry, instead betting they can drift behind American companies as they perform the arduous and expensive research and development to advance the field.

“One of my concerns is that our adversaries just see it as much more beneficial and productive for them as this [U.S. industry] engine burns,” said Harrigan. “Where they actually just see it as more beneficial for themselves in the short term, to hang six to 12 months behind the United States, let us burn really hot and for 5 to 10% of the spend, they can just distill and steal, and they can do that up to until the point where our market kind of falters. At which point they can turn it into high gear, having not expended those resources yet in the past, and actually leapfrog us.”

As part of that mission, NIST will first have to define what a distillation attack is, and separate it from much of the sharing and collaboration between open-weight and open-source models that have also helped drive innovation.

Distillation attacks happen when one AI model is trained on the output data and reasoning of another, usually more powerful or advanced, model. The U.S. government, along with AI companies Anthropic, OpenAI and Google, has accused Chinese companies like Moonshot AI, DeepSeek and Alibaba, of carrying out “systematic” campaigns to distill and copy U.S. commercial models.

According to AI and cybersecurity experts, these campaigns involve the use of hundreds of burner and resold accounts and false identities to bypass geographic restrictions to access U.S. frontier models, and involve thousands, sometimes millions of prompts meant to extract valuable outputs and capabilities that can be used to train a foreign model.

Meanwhile, U.S. AI companies themselves face multiple ongoing lawsuits from media organizations, creators, artists and other laborers, who allege that commercial U.S. models were originally trained on their works without any compensation or permission.  

“AI is advancing at a breakneck pace, and every month we keep China and other adversaries from stealing our progress is a month we gain to build the safeguards this technology needs,” Rep. Whitesides said in a statement to CyberScoop.

The bill would also establish an “AI Cybersecurity Roundtable” at NIST composed of governments and AI industry stakeholders that would work to develop voluntary standards for defensive use of AI in cybersecurity in critical infrastructure.

Harrigan said he envisions the roundtable as including frontier AI companies including software and hardware security experts, who can add independent third-party voices in discussions on the best way to defend IT assets in critical infrastructure and other industries. A place “where you can go and pick an off-the-shelf set of standards that says these are the best practices as we know them all today, given the current state of technology development and the threats that are out there.”

The bill defines “entities of concern” as any foreign person or organization headquartered in one of the four listed countries of concern (Russia, China, Iran and North Korea), or one operating under the direction of a company based in those countries. The director of NIST would also have the ability to designate specific individuals or organizations as entities of concern.

The post Exclusive: Bipartisan bill would develop federal guidance for distillation attacks, convene industry on defensive cyber appeared first on CyberScoop.

The federal government is advising AI companies to set more industry standards for securing agents in sandbox testing environments.

In joint guidance from the Cybersecurity and Infrastructure Security Agency, the National Security Agency and the Center for Advancing Innovation and Standards for Super Intelligence (CAISSI) at the Department of Commerce, AI companies are encouraged to lean more heavily on cybersecurity technical controls, independent testing and evaluation of new models and develop “shut down policies” that govern when a model or test should be taken offline.

The guidance comes after a summer when OpenAI, Anthropic, Google, Meta and other companies say their AI models escaped from testing environments and hacked real organizations and websites. The problem has bedeviled frontier AI companies, with many of their own researchers saying it’s possible that hundreds – even thousands – of similar sandbox escapes may have occurred over the summer, raising major questions about the lack of effective monitoring in place.

“While these incidents highlight the importance of traditional cybersecurity principles and practices, they also introduce novel considerations for incident prevention, response, and reporting,” the agencies wrote. “These novel considerations primarily stem from the agentic nature and advanced cyber capabilities of cutting-edge SI systems.”

While voluntary, the document marks the federal government’s most detailed cybersecurity guidance for securing sandbox environments during AI model tests. 

The first set of recommendations deals with designing and implementing technical controls for the virtualized testing environments, while another recommends outside monitoring tools to ensure AI models don’t access tools or privileges they shouldn’t have. Cybersecurity experts have said both features were lacking in the testing environments used by AI companies this past summer.

AI testers must also scour their internal testing environments for “potential persistence mechanisms and unauthorized communication across internal and external systems” that could be used by AI agents to leave notes and collaborate on tasks in secret.

It specifically cautions against relying solely on other AI systems to monitor and restrain the behavior of their test models, using the government’s now-preferred term of “Super Intelligence” or “SI.”

“Test and red-team SI control measures with humans and SI systems before deploying them,” the guidance states. “Only use SI systems to red-team in cases where out-of-scope actions could not result in harm to critical systems.”

Security teams must evaluate risk not only from the large language model, but also from the surrounding systems and infrastructure it relies on, like the technical harness, the host and broader network, and each layer has security protocols that can stop unauthorized access to the internet or detect anomalous behavior.

Another recommendation calls for frontier AI companies to be more open to partnering with “independent, external auditors” to assess their controls and monitoring for AI agent testing. Critics have argued that Anthropic and OpenAI rely too heavily on a small group of little-known nonprofits and private research organizations, like METR and Redwood Research, to test their frontier models for safety and security concerns and respond to incidents like the Hugging Face hack. They question why these companies don’t work with more established cybersecurity or incident response firms.

Chris Painter, METR’s president, told Congress earlier this month that it had become industry practice for frontier companies to rely almost entirely on other AI systems to monitor and restrain their models in testing environments, citing the sheer volume of data and telemetry that researchers must sift through.

Cybersecurity and AI experts have told CyberScoop that a combination of human oversight, technical controls and AI systems is the most effective means of restraining the kind of agent hacks that have consumed the news.

The guidance further calls for executive leadership at these companies to be more personally involved in decisions around pausing or restarting a particular project, and only after they’ve been informed of the relevant potential risks.

The post Federal agencies recommend cyber controls, independent testing for frontier AI systems appeared first on CyberScoop.

Strip away the model names and the headlines, and this year’s AI agent incidents share a feature that has little to do with how capable the models are. Nearly all of them ran through the software supply chain.

Start with the case that set off the summer’s wave of disclosures. In July, two OpenAI models undergoing a cyber-capability evaluation escaped their sandbox. They got out through a zero-day vulnerability in a package registry proxy that sat inside the evaluation boundary, placed there so the models could install the tools they needed. From there, they reached the open internet and went after Hugging Face’s production systems, looking for the benchmark’s answer key. Hugging Face caught and contained the intrusion on July 16. Five days later, OpenAI confirmed the incident and connected it to an internal cybersecurity evaluation of frontier AI models.

Soon after, Anthropic disclosed three incidents in which its Claude AI models compromised external organizations after a misconfiguration left live internet access in place. In September, it added a fourth that a wider review had turned up. The incident that Anthropic says worries it the most involved a model publishing malicious packages to the Python Package Index (PyPI).

Separately, researchers have attributed more than 2,000 RubyGems uploads in May to OpenAI agents. OpenAI says its agents were carrying out benign tasks. RubyGems removed more than 500 of the packages as malicious.

In each case, the agent’s route out, or the harm it caused, ran through a package registry, a dependency or a model file.

None of this surprised the researchers I work with. ReversingLabs has spent years taking apart the files that move through software pipelines, and we watched attackers discover the AI supply chain well before agents reached production. In early 2025, our team found malicious models on Hugging Face whose pickle payloads were packed in a 7z archive that Picklescan, the platform’s scanner, could not open. The payload itself was an ordinary reverse shell. Hugging Face took the models down after we reported them.

In April, we published research on PromptMink, a campaign we tie to the North Korean group Famous Chollima. In one case, a Claude Opus-based coding agent added a malicious dependency to a cryptocurrency trading project. What struck our team was who the attackers were writing for. Their package documentation was built to be more convincing to an AI coding assistant than to a human developer.

Put those cases side by side, and the threat model looks different from the one most enterprises are planning around. The evaluation escapes show agents misusing the access they were given. PromptMink shows agents being steered by content written specifically to fool them. In both, the damage moved through software the agent downloaded or published.

That is the context for the Open Agent Safety Platform NVIDIA launched on Sept. 28. The design starts from the right premise: you cannot secure an agent by asking it to behave. OpenShell runs each agent in a sandbox and applies policy outside the agent’s own process. Sentry, part of a reference design built on BlueField-4 data processing units (DPUs), watches agent activity from separate hardware, out of reach of a compromised host. More than 100 partners signed on at launch, ReversingLabs among them.

The Hugging Face incident shows why that architecture matters. OpenAI ran those models with cyber refusals reduced, meaning their built-in guardrails were deliberatelyweakened. That left the sandbox boundary as the main line of defense. Prompts and training shape what an agent attempts. Infrastructure decides what it can actually do.

It also matches what I hear from customers. They are demanding AI safety that closes as many gaps as possible and uses the boundary between hardware and software to deliver security they can rely on. That is the standard every vendor in this ecosystem, mine included, will be measured against.

Look, though, at where that boundary failed: a package proxy, trusted because it was supposed to be there. A policy engine can tell you an agent is permitted to install a package. It cannot tell you whether the package is what it claims to be. NVIDIA’s CISO, Saša Zdjelar, made the point himself in a Sept. 21 post, writing that teams must “verify the source and integrity of the tools, skills and dependencies agents use.” Doing that means inspecting the artifact itself, whether a model file, an archive, a container or a binary, at the boundary and before the agent touches it. Even AIUC-1, a prominent certification standard for AI agents, treats scanning model files for malicious payloads as a supplemental control rather than a required one. Its vendor due diligence requirement stops at the model provider. That is the layer ReversingLabs brings to the platform.

I have two cautions as Sentry moves from reference design to shipping products. The first is about openness. Sentry depends on specific NVIDIA hardware, and “open” will only mean something if enterprises can plug in the inspection, telemetry and policy tools they choose and run them on infrastructure they already own. Buyers should hold the ecosystem to that standard.

The second is about the need for prompt disclosure, and it applies to the whole industry. In June, an OpenAI agent accessed non-public files on the Australian government’s Medicare Statistics Reporting Service portal. Services Australia, which runs the portal, learned of it on Sept. 10, through an email to a public mailbox Prime Minister Anthony Albanese called the delay unacceptable. RubyGems maintainers learned who was likely behind their May flood of packages only from outside researchers, months after the fact.

We would not accept that timeline for any other security incident, and AI agents should not be an exception. Organizations that run agents, AI labs included, should notify affected parties on the schedules we already expect for breaches. Package registries are shared infrastructure for the entire software economy, and they should be told when an agent has published to them. Labs running offensive evaluations should treat every service reachable from the sandbox, the package proxy included, as an attack surface, and they should publish the containment standards they hold themselves to.

Compromises like SolarWinds, 3CX DesktopApp and xz Utils taught the security industry that software cannot be trusted simply because of its reputation or where it came from. Agents now pull and publish that software on their own, at a volume no review team can match. NVIDIA has built a strong platform for enforcing trust decisions. The rest of us have to make sure those decisions rest on knowing what is inside the software.

The post Your AI agent is only as safe as the packages it downloads appeared first on CyberScoop.

Anthropic announced a new program Thursday that will combine its AI tools with outside cybersecurity companies to find and fix cybersecurity vulnerabilities in critical infrastructure and open-source software.

Advertised as part of a new, “long-term commitment” to cybersecurity, the critical infrastructure defense program will pair Claude models, Anthropic engineers and threat research with the expertise of cybersecurity companies, including Accenture, Booz Allen, CrowdStrike, Deloitte, Dragos, Hitachi, Insane Cyber, Nozomi Networks, Palo Alto Networks, PwC, and Rockwell Automation.

“Defenders of critical infrastructure and the [open-source software] community have decades of security experience but have faced severe resource shortages that are exacerbated by this moment,” Anthropic wrote in a blog post. “We are recognizing their expertise in these domains and offering our support.”

Anthropic said it has offered frontier models and technical support to more than half the states in the U.S., as well as large operators of critical infrastructure to scan and patch code or assist in incident response and red teaming activities.

“Critical infrastructure is hard to defend in many ways that AI cannot fix, but we believe that frontier models can help find and repair weaknesses before those weaknesses are used to cut off power or make water unsafe, the company wrote. “Our first step is to work with a small cohort of providers to learn which strategies are most effective and practical.”

The company said it has engaged in similar work with maintainers of large open-source software projects, and some organizations asked for everything the model had found in their software, even unreviewed findings.

That spurred Anthropic to create a new opt-in scanning service for open source software, where organizations can get free, periodic scans of their projects with a proof of concept, explanation and suggested patching options. Anthropic noted that under this program, organizations will receive their reports faster but they may contain inaccuracies.

The long-term goal is to automate most triage and patching and develop new security architectures and coding standards.

As AI models have developed increasingly powerful cybersecurity capabilities, such as finding and exploiting known vulnerabilities, frontier AI companies have worried that bad actors could gain access to the tools faster than legitimate organizations. Anthropic and OpenAI have since both started programs funneling their tech to businesses and governments, free of charge, for defensive cybersecurity.

The post Anthropic rolls out program for ‘long-term commitment’ to secure critical infrastructure, open source software appeared first on CyberScoop.

As AI agents escaping testing sandboxes and hacking organizations have gone from unprecedented to seemingly routine in a matter of weeks, policymakers, regulators and cybersecurity attorneys largely agree on one thing: Something must be done to hold AI companies accountable for the incidents.

Exactly what can be done under our current laws and regulations is much less clear.

The conversation has moved from policy and industry chatter to the floor where the future of liability will be determined. Speaking about the Hugging Face hack at a Senate hearing this week, Georgetown University law professor Paul Ohm summarized the argument.

“If you take any of the lengthy reports that have summarized what happened at OpenAI in July and August, and you simply search for the words ‘AI agent’ and you replace them with the words ‘OpenAI employee,’ the document you would be left with would read like a criminal indictment containing the defendant’s own confession of guilt,” said Ohm.

CyberScoop spoke with members of Congress, former federal law enforcement officials, cybersecurity attorneys and other experts about which laws, regulations or policies might apply to agentic hacks carried out by models at Anthropic, OpenAI, Meta, Google and other companies.

Their answers varied widely. Some endorsed using existing laws like the Computer Fraud and Abuse Act, while others said regulators such as the Federal Trade Commission could investigate and fine AI model developers by defining unauthorized agentic hacks as a form of unfair or deceptive trade practice. Still others pointed to civil lawsuits, new laws or state regulators as ways to hold AI companies or users accountable.

But, similar to a familiar technical cybersecurity idiom, there are no silver bullets in this debate.  Nearly every option has real complications or roadblocks that could limit its effectiveness. 

For once, the CFAA is too narrow

One of the first laws mentioned in discussions about agentic AI hacks is the Computer Fraud and Abuse Act, the federal government’s primary criminal hacking law. Historically, many in the cybersecurity community have criticized the statute as overly broad.

For agentic hacks, the problem is the opposite: The CFAA’s language does not clearly cover the activity involved in incidents like the Hugging Face hack.

“I would not be looking at a CFAA charge as the statute exists today” for these hacks, said Leonard Bailey, former head of the cybersecurity unit in the Computer Crime and Intellectual Property section at the Department of Justice.

Bailey helped develop policies that moved the DOJ away from prosecuting “good faith” third-party security research under the CFAA. He said that even if Congress changed the law to cover such hacks, it’s unclear whether that would help since laws like the CFAA are supposed to be “technology neutral.”

According to the Department of Justice’s website, CFAA prosecutions must prove that a defendant accessed a computer “without authorization” or in a way that “exceeds authorized access.”

Crucially, the law specifies that “as part of proving that the defendant acted knowingly or intentionally, the attorney for the government must be prepared to prove that the defendant was aware of the facts that made the defendant’s access unauthorized at the time of the defendant’s conduct.”

In other words, prosecutors must prove that the unauthorized access was intentional, not accidental.

Applied to incidents like the Hugging Face hack, the legal problem becomes clearer. If a human performed the same actions, they would almost certainly face CFAA charges. If a bot or software acted on behalf of a cybercriminal or group, the individuals or organizations intending to profit from the scheme could also face charges.

But no human at OpenAI, Anthropic or another frontier AI company directed, asked or even suggested that its agents hack victims or commit crimes. If charged, those companies would almost certainly argue that none of their actions demonstrated an overt attempt to commit a crime or authorize access to systems or data without permission.

However, others said there may be room to argue that AI companies are now fully aware that their products can engage in unauthorized agentic hacks.

“I’m of the opinion, because it has already happened [and] has happened several times, the argument that it happening again is not at a minimum knowing, for me it’s unlikely because we already have a situation where we know the capabilities,” said Elimu Kajunju, a privacy, cybersecurity, and AI governance attorney at Rimon Law. “The very first one where this happened, that’s the only organization that could say ‘You know what, we didn’t know it could do that or act like that.’ Once we’ve had a second or third or fourth, we can’t say that anymore.”

Kajunju acknowledged that applying statutes like the CFAA to agentic hacks could be complicated. But he said it would align with the spirit of American jurisprudence, in which businesses are often legally accountable for damage they cause, whether they intended it or not.

“I think this is one of those situations where, ‘Does it fit neatly into one of the buckets [we have]?’ No, but can you find one that will fit depending on what has happened? I think so.”

CyberScoop has reached out to OpenAI, Anthropic and Google for comment.

The FTC, states and civil lawsuits

Beyond criminal law, the hacks could also draw scrutiny from federal regulators, including the FTC.

Both Bailey and Kajunju pointed to the FTC as a possible regulator for agentic hacking incidents if the agency defines them as unfair and deceptive trade practices under Section 5 of the FTC Act. One benefit of relying on regulators such as the FTC is that they can often move faster on investigations and enforcement actions than the DOJ can on criminal prosecutions — an important consideration in the fast-moving AI space.

The FTC recently confirmed it was investigating OpenAI, Anthropic and other frontier AI companies, news first reported by The New York Post and confirmed by Axios. The FTC did not return a phone call from CyberScoop requesting comment. 

But Bailey warned that without a specific congressional mandate directing the FTC to regulate AI company hacks, any effort by the agency to redefine or expand its rules would almost certainly be challenged in court.

Civil lawsuits and state policymaking offer additional paths to investigating or regulating AI companies.

Currently, Florida is investigating OpenAI over the HuggingFace hack, while a nonprofit that sued OpenAI earlier this week has cited alleged violations of California law.

In his testimony, Ohm said that if Congress was serious about preventing agentic hacks, it should not pursue federal legislation preempting state AI laws. He called states “laboratories of democracy,” that can experiment with different forms of regulation and lawmaking.

Others agreed with that approach.

“Obviously this is a topic that the federal government doesn’t love, states getting involved in this space, but it just takes too long for action to come through Congress,” said Kajunju. “I think the quickest way to getting us to a better place will be a really good state law.”

New legislative remedies

On the Hill, members of Congress are working through the same questions as they consider legislative remedies.

At a Senate Homeland Security Committee hearing Wednesday, Sen. Josh Hawley, R-Mo., took a minute at the outset to describe the volume of agentic hacks affecting outside entities that frontier companies appeared to have missed for months. He named affected targets one by one, including major online code repositories like HuggingFace, Australian and U.S. government websites, obscure foreign-language wiki sites and other parts of the internet.

After weeks of investigations uncovered additional, previously unreported incidents in which models escaped testing environments and hacked victims, Hawley said it was “time to have a conversation about who bears responsibility” for the fallout. He added, “as soon as the AI executives say ‘we got a real problem here, our product is out of control,’ the very next words out of their mouths are, ‘but we’re not responsible for it.’”

Hawley said the committee invited OpenAI CEO Sam Altman to testify at the hearing, but he declined. Hawley then accused frontier AI companies of emphasizing their products’ danger while pushing for legal antitrust exemptions and only supporting government action if “they can write all the regulations.”

“I’d suggest that maybe we start with some federal legislation…and we can use existing statutes to do it, I propose using the Computer Fraud and Abuse Act,” said Hawley. “It’s already on the books, we can just update it to say that for developers – that’s the companies – if you develop these agents and train them in a reckless fashion and they go on to hack and destroy stuff, you’re liable.” 

Like other legal experts, Hawley said he believes a CFAA prosecution requires a human actor or organization that intended to commit a crime.

Another Senator, Ron Wyden, D-Ore., said he was also working on legislation related to the hacks. 

“The big AI companies should be held responsible if their agents hack other companies, government systems or foreign websites,” said Wyden in a statement to CyberScoop. “In my view there’s a question about whether the current CFAA would apply to the recent hacks. I’m working on a narrow update to the law to ensure Anthropic, OpenAI and other big AI companies can be punished when their agents run wild.”

Sens. Mark Warner, D-Va., Brian Schatz, D-Hawaii, and Andy Kim, D-N.J., introduced a bill that would create an AI Safety Board at the Department of Commerce. That board would evaluate emerging AI risks and establish safety and technical standards.

The Trump administration’s testing regime with AI companies is entirely voluntary, applies to only certain models, and requires testing no more than 30 days before release. Under the Democratic bill, frontier AI companies would be forced to submit models for testing 45 days before release. They would also be legally required to comply with government safeguards related to models discovering and exploiting software vulnerabilities without explicit human direction. Companies could face fines of up to $250,000 per violation, per day.

When asked whether the CFAA or other federal statutes could or should be used to prosecute companies when their AI systems hack, Warner said that companies’ claimed lack of intent to harm should not become a catchall legal excuse for avoiding responsibility. AI companies have said they never directly prompted or suggested that their models hack or commit crimes.

“AI models are tools and if I buy a power tool and deliberately misuse it, that’s my responsibility,” said Warner. “But if a defect causes that tool to behave in a dangerous way, we don’t absolve the manufacturer simply because someone else was using the tool when it malfunctioned. As AI systems become capable of acting autonomously, we need to ask what principles of responsibility should apply to the companies that design and deploy them.”

The post The legal questions raised by agentic AI hacks appeared first on CyberScoop.

Collaboration with industry is key to balancing AI security risks and benefits, as well as staying ahead of China and other adversarial nations, National Cyber Director Sean Cairncross said Thursday.

The Trump administration is facing pressure from some quarters of Capitol Hill and even some artificial intelligence executives to establish regulations or embrace legislation to guard against the technology’s risks, something the president himself has rejected.

“We’ve never faced a world where our adversaries have the frontier in this space, and so our attempts to optimize that innovation and security is is something that is ongoing, but is increasingly efficient, and it depends on our relationship with industry and our ability to work collaboratively among the the interagency,” Cairncross said at the AI Edge event hosted by The Washington Post.

That collaboration is ever-increasing, he said, such as with the National Institute of Standards and Technology’s Center for Advancing Innovation and Standards for Super Intelligence (CAISSI).

“Industry coordination on this, across critical infrastructure sectors, is improving both on industry-to-government and within industry,” he said. “There is collaboration with CAISSI and others as well to do evaluation and testing on these models. That includes work on guardrailing, sandboxing, et cetera.”

But “we’ve seen in the past, once the government is introduced into this space directly, there is a tendency for the government to start to want to adjust the dials directly, and it is difficult to reverse that,” he said. “I think that would slow, at this point, the innovation cycle that is to our benefit.”

One of the things that has increased the calls for stronger government action is when OpenAI acknowledged that its AI agents went outside testing to hack Hugging Face in July.

“Since that incident,” Cairncross said, “the engineering work that’s gone into improving systems awareness of that has increased by an order of magnitude.”

But the AI companies need to secure themselves against China for the United States to stay in the lead, he said.

“Both on the hardware side and on the computer science side and engineering, we’re in the lead. I think our lead is significant,” Cairncross said. “The Chinese have designed a sort of fast follow that relies quite a bit on distillation efforts … And I think industry is increasingly aware of this: that their systems are secure, that the technology, as it’s being developed, is protected.”

Other administration officials have called out China for allegedly using distillation to train its models. In July, Michael Kratsios, who leads the White House Office of Science and Technology Policy, claimed that Moonshot AI, a Beijing, China-based AI company, had distilled Anthropic’s recently released Fable model to develop its own K3 model.

On Wednesday, OpenAI said clusters of activity tied to Moonshot AI were part of a “coordinated” distillation campaign.

Cairncross said earlier this week that the U.S. government is collaborating with industry to integrate AI into critical infrastructure for security purposes. 

Cairncross said the Trump administration wants to make sure national security systems do the same. One of his office’s pilot programs is with the Defense Department and the Cybersecurity and Infrastructure Security Agency “that touches on the defense infrastructure in particular,” in addition to a pilot on rural hospitals.

The post National cyber director: Government-industry collaboration vital to managing AI risks, competition with nations appeared first on CyberScoop.

Amid growing concerns from both the public and policymakers about cyberattacks involving advanced AI systems, NVIDIA has released a new open software security platform for AI agents.

The Open Agent Safety Platform, according to NVIDIA, will be “an open software platform and reference system design with full-stack governance and control across the software and hardware, compute and robotics systems that run agents.”

More than 100 organizations from the AI industry, including Anthropic, Arm, Microsoft, SpaceXAI, Palantir and JPMorgan Chase, have committed to using the platform to improve security in their products.

Stronger sandboxes and more advanced monitoring are pillars of NVIDIA’s strategy for tackling “rogue” AI agentic hacks, and the platform offers a pair of tools that are meant to help contain and restrict agent activity. 

According to a corresponding technical blog, OpenShell, built on Apache 2.0 open source software, is a tool for securing runtime execution for AI agents in sandbox testing environments. AI system operators can define files, networks, tools, processes and credentials that an agent has, and test whether those guardrails hold before introducing them to enterprise networks.

It also includes security updates for NVIDIA’s Bluefield 4 data processing unit (sometimes called a “data center on a chip”) that enables out-of-band monitoring of AI agent behaviors and security policy enforcement.

On X, NVIDIA CEO Jensen Huang called the platform “the beginning of an open ecosystem to build the trust layer for safe agent systems.”

To do this, security work must become “foundational” to AI and the industry’s “full promise can only be realized when people have confidence that AI is being built to be safe and deployed with wisdom and responsibility.”

“Trust and innovation are not in conflict,” Huang wrote. “Safety is how trust is earned. We must build not only the most capable AI, but the most trusted AI, so that this extraordinary technology can realize its enormous promise for the world.

NVIDIA manufactures advanced computer chips that power much of the U.S. commercial AI industry. After models from Anthropic, OpenAI, Meta and others escaped sandbox protections during testing and breached real organizations, some AI industry leaders – including Anthropic CEO Dario Amodei and OpenAI CEO Sam Altman –  suggested AI systems are too advanced to be contained.

Huang has criticized those views, arguing that frontier AI companies and their supply chain partners must improve their security posture. Huang said last week he opposes government regulations or mandates on the AI industry, unless they promote growth.

In an interview with CNBC on Monday, Huang said he believes sandboxing and larger AI security issues are “a technically solvable problem.”

“I think the answer is we hope it’s an engineering problem,” said Huang. “I believe it’s an engineering problem, I know it’s an engineering problem, and we all need to hope that it’s an engineering problem. [If not] it’s not solvable, so the fact that all of these companies still are advancing the state of the art is because they also believe it’s solvable.”

Aviv Nahum, CEO of Above Security, told CyberScoop that NVIDIA’s announcement reflects industry recognition that model training alone cannot ensure safety. 

NVIDIA is essentially arguing that for agents to perform safely in real-world environments, “some of the enforcement must live outside the model, in a layer the agent cannot simply reason around or modify.”

“What this announcement says to me is that the industry is finally converging on a basic cybersecurity principle, [that] model alignment is not a substitute for security engineering,” said Nahum. “Sandboxes, identity, least privilege, independent monitoring and containment are not new ideas. What is new is that we now have autonomous software capable enough that failing to apply those principles becomes much more consequential.”

The post As AI world debates security, NVIDIA releases open source tools for agents appeared first on CyberScoop.

A new Democratic bill in Congress would establish a federal Cybersecurity and AI Board of Investigations to provide independent government oversight of cyberattacks carried out by AI agents, following recent hacks by models run at companies like Anthropic, OpenAI, Meta and others.

The bill, introduced by Sen. Ed Markey, D-Mass., would attempt to establish a federal mechanism to investigate incidents where AI models escape sandbox environments and access live internet systems.

Currently, frontier AI companies like OpenAI and Anthropic largely control the investigation and public reporting of such incidents. Markey and other critics argue that these companies have too much control over investigations and reporting due to their financial and legal interests. 

“Despite the unprecedented depth and scale of recent AI-enabled cyberattacks, the public is learning critical details piecemeal,” Markey said in a statement. “Building stronger defenses requires a full accounting of what goes wrong, and we cannot depend on companies with little incentive to disclose their failures to give us one. We need the Cybersecurity and AI Board of Investigations to get to the bottom of major incidents and give companies and the government the critical information necessary to build resilience and better secure our economy and our country.”

Although frontier AI companies maintain external red-teaming programs and allow limited access to organizations like METR and Redwood Research, they control the scope, terms and time frames of those engagements.

The board, which would coordinate with the secretary of commerce, could subpoena witnesses and conduct “independent and impartial reviews and assessments” of AI agent-led hacks that impact federal information systems or critical infrastructure. 

It would be led by five members, appointed by the president and confirmed by the Senate for five-year terms, with no more than three members from one political party.

The board would also investigate systemic vulnerabilities in the AI supply chain, so-called “near misses” where unauthorized agent-led hacks were “narrowly averted,” and gaps in federal regulatory oversight. It would have technical staff including engineers, malware analysts, and digital forensic experts.

The board would “operate independently from regulatory review and enforcement actions without assigning legal fault or liability for any review and assessment” it conducts, according to the bill.

OpenAI confirmed Wednesday its AI agents breached a statistics portal used by the Australian government’s social services agency, Services Australia. Though the breach happened in June, OpenAI learned of the incident in August. Australian Prime Minister Anthony Albanese said the company did not notify him until Sept. 10, when it sent findings to a general government email inbox, according to the BBC.

The post New bill would create federal investigative body for AI-driven hacks  appeared first on CyberScoop.

President Donald Trump continued to defend his administration’s hands-off approach to AI regulation in the wake of hacks carried out by U.S. commercial frontier models that have rattled policymakers and industry veterans and spurred calls for more regulatory oversight.

In a Truth Social post Monday, Trump dismissed worries from critics that “AI is going to kill us,” comparing them to complaints from environmentalists about climate change, which he also alleged was a false narrative. He also posited that nothing may matter more than future U.S. dominance of the technology over geopolitical rivals like China.

“Whoever wins AI, WINS!” Trump posted. “We are leading now over China, and everyone else, and I’m going to keep it that way! I’m not going to stifle Growth, of something that will be bigger than the Industrial Revolution, or the internet, itself.”

Trump has previously suggested that good leadership is the only regulation the U.S. needs for artificial intelligence. He later claimed the Department of Justice was ready to “rein things in” if companies overstepped, but offered no specifics on enforcement, legal authority, or where he would draw that line.

“We will be careful, and that’s why we have the Department of Justice, and other Law Enforcement bodies, that will rein things in if we have to, but I will only encourage AI or, SI (SUPER INTELLIGENCE)!” Trump concluded.

Secretary of the Treasury Scott Bessent recently told Congress that private lawsuits could force AI companies to institute better security, saying it’s clear what the government “shouldn’t do on safety is to give these labs a liability exemption, which is what they are asking for.”

“The best way to guarantee safety is that the creators are liable for what they build and generate,” Bessent said.

Beyond existential fears, critics also argue that inadequate regulation or cybersecurity controls in current AI systems make them impossible to fully control or monitor.

Recently, former President Barack Obama criticized the argument from Trump administration officials that the free market will naturally push industry toward self-regulation and that “these companies will solve the safety issues because they have every incentive to do so.”

“If it turns out to be dangerous, people will just sue them and they’ll be worried about financial liability,” Obama said last week in remarks at Colgate University in New York. “That’s not how we treat airlines or drug companies or food companies.”

The Trump administration issued an executive order earlier this year that set up a voluntary testing regime for some commercial frontier models, largely at private industry’s discretion. That order was significantly delayed and altered by AI industry boosters to ensure that governmental review did not cause companies to postpone their release timelines for new models.

That agreement did not last long before fast-moving events caused the administration to strike another, non-public agreement with frontier AI companies like OpenAI, Anthropic and others governing pre-release testing for models.

But the Trump administration has consistently argued that regulation will harm, not help, U.S. innovation and global competitiveness, and the threat of China frequently looms large in those discussions.

Experts believe China’s AI models are behind U.S. models at the top of the market, where OpenAI and Anthropic have consistently pushed the frontier limits of model capabilities. But Chinese lower and “middle class” models are often cheaper, more efficient and can even outperform more powerful models because users can dedicate exponentially more tokens for their tasks.

The U.S. government has accused Chinese AI companies of conducting widespread, “systematic” distillation of U.S. frontier models, with the implicit encouragement of Beijing.

In defending the administration’s approach, David Sacks, co-chair of the President’s Council of Advisors on Science & Technology and a top adviser on AI issues, specifically cited the threat from China and other countries that he claimed would not be subject to similar restrictions.

“We’re not the only country that has advanced AI labs, and as the president declared…we have to win this AI race,” Sacks told Politico in May, later adding “I think that’s the first thing to recognize is that if somehow we slow down or stop AI development, it doesn’t mean that AI progress is going to stop. It just means it’s going to happen in other countries and specifically China.”

Some observers have alleged that despite their larger differences, top leaders in the U.S. and China may view AI similarly at the strategic level, specfically that increased adoption – and risks – of AI are inevitable.

Ronan Murphy, director of the tech policy program at the Center for European Policy Analysis, posited that while there may not be a formal agreement between the two countries, “they share views both in Beijing and in Washington, particularly in the White House, of: you have to allow this to happen.”

“Clearly there’s a call for regulation from many quarters of AI in the U.S. and elsewhere, but in the White House – and we heard David Sacks talking about it [recently] – It’s ‘let them cook,’ and the Chinese approach seems to be the same,” said Murphy in a press briefing. “So there might be consensus at that level, if nothing else.”

The post Citing China, President Trump doubles down on hands-off approach to AI regulation appeared first on CyberScoop.

Researchers said they used Anthropic’s Claude and OpenAI’s Codex to identify a damaging flaw embedded in a popular software decoding tool that could leave major internet platforms, enterprise services, and web frameworks vulnerable to data theft and remote access.

The vulnerability, nicknamed HEIF Heist, refers to the malware’s ability to trigger memory corruption errors in affected software, allowing the attacker to pilfer sensitive data from its victims. In a report published Thursday, the researchers laid out the potential damage an attacker could cause, including gaining access to internal OpenAI repositories, leaking user files, access tokens, and other sensitive data for online services like Amazon Web Services, and gaining remote code execution privileges across a range of online services, including Meta’s core product suite, GitHub Enterprise servers and open-source internet forum Discourse.

“Even when Remote Code Execution isn’t immediately achievable, the attack primitives may still allow arbitrary heap disclosure, letting an attacker ‘heist’ in-memory data such as other users’ data and environment variables,” wrote Hacktron researchers Harsh Jaiswal, Mohan SRK, Rahul Maini and Sudhanshu Rajbhar.

The researchers relied heavily on AI systems, including frontier models from OpenAI and Anthropic, to conduct their research. Attribution for the research is described as being “led” by the Hacktron human researchers “assisted by Hacktron Harness, GPT-5.6 Sol, and Opus 5.”

According to the research, the attack exploited the way that code parsing tools in many popular software decoders — specifically libheif and libde265, used to parse C and C++ software — process certain image files.

By uploading HEIF, HEIC and AVIF image files corrupted with malicious code, the attacker could bypass most of the victim’s application layer defenses, in many cases achieving remote code execution privileges for accounts or products tied to major AI and tech brands.   

While the latest version of libheif has been patched, the researchers said “any deployment lacking the latest upstream security patches is potentially vulnerable.”

In one incident detailed in a Sept. 13 blog, Jaiswal, Maini, and Hacktron researcher Mohan Pedhapati described how chaining two vulnerabilities, including an image parser flaw, could compromise OpenAI employee accounts.

With access to the compromised accounts, researchers could reach OpenAI’s internal repositories. As a proof of concept, they opened a pull request in the company’s “monorepo,” a centralized library where code is shared across projects, using the employee’s Codex credentials. 

According to a timeline provided by the researchers, the flaw was discovered on July 25 and patched within days. They said the entire attack, from discovering the initial vulnerability to gaining access to the repositories, took less than 72 hours. OpenAI paid them a bug bounty of $6,500 for their work.

Given that AI models are increasingly integrated into enterprise and personal networks, an attacker exploiting HEIF Heist could have accessed far more than just OpenAI’s systems and data.

“Until two months ago, a user or OpenAI employee logging into OpenAI’s own help forum could have had their ChatGPT and Codex accounts taken over,” the researchers wrote. “Since people can connect various services to Codex and ChatGPT, the scope of what we could theoretically access was huge, including GitHub, Slack and emails.”

CyberScoop has reached out to OpenAI for comment on the research and additional information.

At the same time, the researchers said the attack paths they found were not particularly easy or efficient to exploit.

“Exploitation requires fingerprinting the target version and tailoring the payload images,” the blog stated. “Some of our RCE attempts landed only after thousands of image uploads. That said, an AI agentic approach with a frontier model like GPT-5.6 Sol cut exploit development time down to roughly 1 to 3 days from initial probe to remote RCE. A motivated attacker can convert a vulnerable upload endpoint into RCE or an info leak.”

The post Researchers use AI to find widespread software decoder flaw  appeared first on CyberScoop.

The past few weeks have “felt very strange” for Juan Andres Guerrero-Saade.

Like many, he is trying to sort through the spate of frontier-model AI agents from OpenAI, Anthropic, Meta and others hacking their way onto the open internet over the past few months, particularly amid the already-heated national debate around the emerging technology and its impact on society.

Guerrero-Saade, a fellow for AI and security research at SentinelOne and an adjunct professor at Johns Hopkins University, said the hacks are worth taking seriously, but at a time when businesses and open-source maintainers should be focused on further hardening their systems and policymakers should be discussing new solutions,  “what we see is cybersecurity being used essentially as an excuse for these AI doomer arguments.”

The incidents have spawned those “doomer arguments” amid an intense public debate about the technology, the pace of industry development, and whether government and the private sector are doing enough to protect against “doomsday”-type scenarios, where AI systems take over or attack large parts of the internet or society.

Guerrero-Saade is among a growing chorus of cybersecurity professionals who say that while AI systems pose real, unique threats to our systems, the apocalypse is far from inevitable. Most of the public concerns around the incidents, let alone worries about killer AIs attacking critical infrastructure, assuming control of the internet and wiping out humanity, are either technically impossible or can largely be controlled through established cybersecurity principles.

There is this “narrative or magical thinking of ‘Well, AI is going to be able to hack everything, and therefore it can control everything, and therefore it’s going to kill us all,’” he told CyberScoop. “And you [think] these just don’t add up. They’re not very well-reasoned arguments.”

This fatalistic narrative tied to AI’s eventual dominance doesn’t hold up under scrutiny, according to experts CyberScoop spoke with. In recent conversations, cybersecurity and national security professionals raised questions about both the technical solutions OpenAI and Anthropic use to contain their models, as well as the glaring absence of federal oversight from federal regulators or truly independent third-party review.

For example, Jacob Coxon, an Anthropic employee who resigned over AI safety concerns, told CBS News that frontier models could not be “unplugged” by humans once deployed because the model would copy itself to thousands of other computers connected to the internet.

By contrast, Matt Tait, a former information security specialist at UK signals intelligence agency Government Communications Headquarters (GCHQ), pointed out that the models run by Anthropic and other frontier companies require extremely expensive, “ultraspecialist” machines that “are functionally supercomputers.”

“There is a zero chance that Anthropic’s most capable models will be able to extract their own model and run in the wild, because those supercomputers essentially only exist in datacenters,” Tait said.

“Not a credible warning”

Other former cybersecurity government leaders say the agentic hacks represent a failure by regulators and industry to deploy known technical and policy options that make it harder for these types of incidents to occur.

Matt Hartman, former deputy executive assistant director for cybersecurity at the Cybersecurity and Infrastructure Security Agency, said “we should not accept harmful AI behavior as inevitable or unmanageable.”

“There are meaningful steps companies can take to monitor agent activity, constrain permissions, detect anomalous behavior, and build stronger safeguards into how these systems operate,” said Hartman, now a chief strategy officer at Merlin Group. “Those controls will inevitably involve trade-offs in capability and speed, but that’s a familiar cybersecurity challenge. Our goal should be to manage the risk without unnecessarily limiting the enormous benefits AI can provide.”

Ciaran Martin, former head of the UK’s National Cyber Security Centre, took issue with the way the CEOs of frontier AI companies have framed the threat of “rogue” AI behavior as inevitable, while issuing dire warnings about future threats and capabilities with little transparency.

Martin’s comments came after an essay published by Anthropic CEO Dario Amodei that cited the threat of a HuggingFace-style swarm of agents that could create a botnet capable of “taking over the entire internet” within 6-12 months.

This, Martin said, “is not a credible warning,” because it doesn’t explain how the exploitation would function, how such a botnet would persist on the internet, or how it would escape law enforcement. 

 “It assumes no monitoring of systems, no anti-virus, no DDoS protection, no network segmentation, no incident management, no nothing of any kind of the cybersecurity on the global Internet of the type that has developed over the last 30 years,” wrote Martin. “For a claim of this magnitude, there is neither evidence for the contention nor a credible account of a path to this outcome.”

Meanwhile, some federal government cybersecurity leaders have touted the technology’s disruptive potential and called for more widespread adoption of AI tools by defenders.

Joseph Alm, assistant secretary of cyber, infrastructure and risk resilience at the Department of Homeland Security, said classified systems may retain stronger protections. But for most other data, AI models are “just going to know things and be able to infer things about the world, and we’re going to have to adapt to that as almost inevitable.”

Asked by CyberScoop whether the government or frontier AI companies could be doing more to prevent or deter their models from carrying out unauthorized hacks via agents, Alm cited recent efforts by the Trump administration this year to establish pre-release testing of commercial models as a step in the right direction. But he called unauthorized AI agent hacks “a new threat class” that is different from previous threats and can be easily distributed to users through open-source software today.

“I think what we can do is…encourage the building of good sandboxes, so that the best models aren’t used for this and the stuff you see out in the wild is the kind of detritus that you can actually respond to effectively and control your networks,” said Alm.

Other experts have shared similar concerns. Earlier this month, CrowdStrike CEO George Kurtz recently warned of a new threat class emerging alongside nation-states, cybercriminals, and hacktivists: “the agent state.” By pairing AI systems with small human teams, these operators can now match the speed, scale, and sophistication of government-backed hackers.

“It took a nation to fund the talent, the tooling, the infrastructure, the patience,” said Kurtz. “That scarcity is over.” 

To be sure, frontier AI companies tout their commitment to both approaches. OpenAI and Anthropic have rolled out an array of cybersecurity partnerships, external red-teaming programs, vulnerability disclosure programs and cybersecurity technical advisory bodies filled with cybersecurity experts.

Mohammed Husain, strategic delivery lead for government at OpenAI, told CyberScoop that the company deploys both internal safety guardrails for their models and relies on outside cybersecurity vendors for additional expertise.

Internally, OpenAI focuses on vulnerabilities at the training level: filtering data poisoning attacks, blocking harmful datasets, and using network controls to prevent prompt injections. For other security layers like sandboxing, identity management, networking controls, they outsource to external vendors. 

“I don’t think OpenAI has all the answers here but what we do as a research lab is we’re going to focus on levels of protection we have expertise in and we partner to self-complement,” said Husain.

AI safety vs. AI cybersecurity

In response to the HuggingFace hack, OpenAI and Anthropic have allowed third-party organizations, such as nonprofit AI research firms METR and Redwood Research, to investigate. But multiple cybersecurity professionals told CyberScoop that both firms lack incident response experience and focus primarily on AI alignment and safety. Their reporting on the hack also lacked critical details: network monitoring logs, telemetry, and other data standard in cybersecurity threat intelligence reports.  

METR president Chris Painter addressed those general concerns in a post on X, saying since 2022 the organization has worked with Google, Anthropic, OpenAI, Meta, Amazon and others on investigations and third-party evaluations. Painter said none of the AI companies fund METR and that his employees are not uniformly “doomer” or “accelerationist” around AI.

Painter also said METR’s work ensures that if AI systems become autonomous or “rogue” within a company, there are ways to share that information with governments and people “outside the company’s walls.”

“We don’t accept money from frontier AI companies,” wrote Painter. “They haven’t paid us for our work, and we don’t accept donations from them or their employees. As we’ve shared previously, multiple frontier AI companies currently provide us with free access to their models in order to perform our evaluations, research, and engineering.”

AI safety and AI cybersecurity advocates take different approaches to securing “rogue” AI behavior. Safety advocates focus on aligning models around ethical training and behavior. Cybersecurity advocates argue that technical and regulatory controls must go further—actively preventing models from accessing what they need to carry out malicious behavior.

Guerrero-Saade said sandboxes in particular can easily be programmed with aggressive cybersecurity monitoring in order to spot when something odd may be happening and react in real time.

“I can’t think of an easier situation in which to set up trip wires, set up configurations like DNS servers, just different parts where you can say ‘Hey, anomalous behavior is happening,’” he said. “We should have been able to tell this immediately, not weeks and months later. So watching [the AI hacking incidents] go down is a little ‘crazy-making’ because we’re seeing things that, frankly, look like neglect, negligence, people just mishandling things, and then being told that these are categorically new incidents that mean that AI systems need to be treated completely different from anything that’s come before.”

While cybersecurity experts say AI systems are, at their core, still software, they do operate differently from more traditional code in ways that can make them harder to predict and control.

John Hultquist, chief analyst at Google’s Threat Intelligence Group, said most software has been deterministic. It may have bugs or vulnerabilities, but an expert could generally understand how it would react to certain stimuli, making it easier to design straightforward controls.

AI models are non-deterministic, with far more variability than traditional software. That can break security controls that rely too much on predicting behavior in advance. Using AI to enforce security controls on other AI models faces the same problem: the systems being deployed to control AI are just as unpredictable. 

But people are also non-deterministic, and people have developed systems in other industries and practices to account for that.

Hultquist drew on his Army experience, noting that “they give incredibly dangerous, expensive things to 18-year-olds” and expect responsible use. The military manages this through two types of controls: deterministic ones like strict weapons and ammunition protocols, and non-deterministic ones like human officers who monitor and correct violations.

Similarly, established cybersecurity controls have been used by incident responders to detect and prevent or mitigate ongoing cybersecurity breaches.

“I don’t think we should throw out all the other tools that we have learned to use as well. I think that would be utterly foolish,” he said, later adding “I will say that if we use only non-deterministic tools to figure out when things are happening, we shouldn’t be surprised when we get the wrong answer.”

The post The AI hacking apocalypse is not inevitable appeared first on CyberScoop.

Artificial intelligence has removed the skill advantage that once set state-sponsored hackers apart from lone criminals, according to a threat report Anthropic published Thursday that documents misuse of its Claude models across seven areas of harm.

The report, which details activity observed between December 2025 and August 2026, covers cyber operations, influence operations, surveillance, scams and fraud, biological misuse, conventional weapons development and distillation. Anthropic said it disrupted each operation, strengthened safeguards and shared intelligence with authorities and industry partners where appropriate. 

“The cases we share here aren’t typical misuse, but rather examples of the most notable and novel threat activity we’ve identified to date,” the report reads. “We’re publishing this work because we believe we have a responsibility to disclose malicious misuse of our services. As models become increasingly capable, their risks will increase, unless AI developers and society’s defenders act to make them safer.”

The cyber operations the company detailed were a Russian-aligned espionage campaign that hit more than 20 government and defense organizations across Ukraine and Europe, two Chinese undergraduates who ran an automated exploit foundry that produced more than a dozen potential zero-days in a single month, affiliates of the ShinyHunters crime collective who dumped 2,100 cloud access tokens across 40 corporate tenants in 34 hours, and a lone hacktivist who targeted European political parties via stolen API keys. 

For decades, cybersecurity researchers and investigators have pointed to sophisticated operations as a signature of state-sponsored tradecraft, while crude intrusions suggested amateurs or petty criminals. Anthropic posits in the report that AI has erased that conventional thinking, especially since a “majority of the operations described in this report were enabled by AI via direct execution or orchestration.”

“For threat intelligence investigators, sophistication has stopped being a reliable signal of who is behind an operation,” the report said, adding that a hacktivist on stolen API keys, scattered criminals and a state espionage operator each ran campaigns that a year earlier “would have required many skilled operators and specialist knowledge.”

The most extensive case involved a malicious actor using the handle “JackPoterz” whose actions aligned with Russian state espionage, matching behaviors linked to Midnight Blizzard. 

According to the report, the actor employed a custom toolkit composed of two families of Windows-based implants, a mobile exploitation kit, a credential stealing tool that targets browser password stores, a phishing platform designed to mimic priority targets like government organizations, and an administrative console used to manage compromised accounts. Targets included military intelligence bodies in Ukrainian and European governments, diplomatic and defense organizations, and people connected to U.S. foreign policy.

According to the report, AI monitored whether security products flagged the actor’s malware. When a detection occurred, “agents would then set about the process of autonomously modifying and rebuilding the malware to evade the existing detections,” the report said.

The same actor bulk-exported mailboxes at drone component manufacturers and stole a complete software development kit for a drone vision system, then spent days recovering its architecture and details of an unannounced product. The actor also compromised hotel Wi-Fi vendors to reach guests through DNS hijacking, took over WhatsApp accounts with headless browsers, and stole more than 300,000 national identity records from a North African government agency, along with registry data on more than half a million companies.

The Chinese-speaking operators, which the company says were partly carried out by undergraduates at a Chinese university, put Claude to work on vulnerability research around the clock. One workflow iterating on network appliance firmware “yielded more than a dozen possible zero day findings in a single month.” It ran “agent swarms,” in which a lead agent divided work among parallel subagents, and kept campaign memory between sessions. 

Clusters linked to ShinyHunters showed how AI shortens criminal timelines. One supply-chain breach ended with a dump of more than 2,100 Azure access tokens spanning more than 40 corporate tenants in about 34 hours. “AI agents performed nearly all of the work,” the report said. Another compromise moved from a single stolen developer token to full control of a victim’s cloud environment in roughly three hours.

The report also has a section dedicated to distillation attacks that Anthropic claims were carried out since February by seven labs based in China, including Alibaba, DeepSeek, Moonshot AI, Xiaomi and Zhipu. Operators affiliated with Alibaba ran the largest attack Anthropic has measured, peaking “at nearly 3 million exchanges per day launched from more than 3,500 fraudulent accounts” to harvest the outputs of Claude Opus models for training its Qwen systems.

The outputs were culled from users who never knew they were involved. The report said Moonshot and DeepSeek silently forwarded their own customers’ requests to Claude and returned its answers as their own, exposing data users had not agreed to share, including surveillance footage of a tracked individual pulled by a user likely affiliated with the People’s Liberation Army. Those practices are “likely inconsistent with privacy laws and the labs’ own terms of service,” the report said.

Earlier this week, a joint cybersecurity advisory from the National Security Agency, the Cybersecurity and Infrastructure Security Agency and the FBI accused Chinese AI companies of engaging in a deliberate and “systematic” effort to illegally distill U.S. frontier AI models and their capabilities.

Anthropic said it published the cases to give outsiders a view of how these threats form, framing the disclosures as an early look at a shifting landscape. 

“As models become increasingly capable, their risks will increase, unless AI developers and society’s defenders act to make them safer,” the report said. The old idea of “security through obscurity,” it added, “is no longer viable in this new AI-assisted world: everything connected to the internet is a potential target for exploitation.”

You can read the full report on Anthropic’s website.

The post AI lets small actors run state-level hacking campaigns, Anthropic report finds appeared first on CyberScoop.

More than 100 companies and organizations, including OpenAI, Anthropic, Google, Microsoft and Amazon Web Services, have signed an open letter calling for a global effort to improve cybersecurity defenses as artificial intelligence capabilities advance.

The letter, published Thursday, argues that the timeframe to strengthen defenses before AI-enabled attacks become more widespread and complex is rapidly dwindling. Conversely, the letter says the same advances can give defenders new ways to find and fix vulnerabilities that have accumulated over years, a period the signatories call a “defenders’ window.”

“Each of us can reduce risk now,” the letter reads. “All organizations, cybersecurity companies, technology partners, governments, and AI frontier companies have an important role: accelerate defenders’ priorities with tools, funding, and hands-on support, especially for critical infrastructure organizations with limited budgets.”

Aside from AI-centric companies, financial institutions like Capital One, Mastercard and Visa, and cybersecurity firms like CrowdStrike, Palo Alto Networks, and Proofpoint, also signed the letter. Organizers describe the effort as ongoing, with more organizations expected to join over time.

The letter states that “status quo security won’t be enough.” It cites longstanding bugs, excessive permissions, misconfigurations, unpatched software, weak authentication and technical debt in legacy systems as sources of exposure. Security teams, particularly those protecting critical infrastructure, have been historically under-resourced, the letter says, and need what it describes as a surge in tools, resources and hands-on support.

In a conversation with CyberScoop Wednesday, top brass from Palo Alto Networks said they had seen enough from internal frontier AI model testing and malicious in-the-wild use of commercially available AI tools to be genuinely concerned.

“I can tell you without exaggeration that we believe that this is a generational shift in cybersecurity,” Sam Rubin, senior vice president of Palo Alto Networks’ threat intelligence arm, said Wednesday. 

John Doyle, CEO of Cape, a privacy-first mobile network operator and whose company signed the letter, echoed the warning about status-quo security.

“It was already failing us in telecom–critical infrastructure that’s been breached time and again with serious consequences for both our military and regular people,” Doyle told CyberScoop. “It’s going to get immeasurably worse without collective action and leaning into innovative cyber defense.”

The letter further asks every organization to make cybersecurity an immediate leadership priority, fix the highest-risk weaknesses and raise security standards for technology they buy, build and deploy, including AI-generated code. Cybersecurity companies and technology partners are asked to test defenses against frontier AI capabilities and make AI-powered defense accessible to critical infrastructure operators. 

Governments are urged to coordinate defense across borders, fund protection for essential services that lack staff or budget, and impose costs on attackers. Frontier AI companies are asked to provide responsible model access, funding, training and support, and to ensure that AI systems acting autonomously remain traceable and accountable.

The letter frames AI as both a threat and a remedy throughout the document. It mirrors what security experts have been saying for months, positioning the industry as entering an unprecedented two- to three-year period of upheaval, driven by AI systems that are discovering vulnerabilities exponentially faster than defenders can respond and threatening to render decades of security practices obsolete.

The U.S. government has taken steps to stay ahead of AI-enabled cyberthreats. As part of an executive order issued by President Donald Trump in June, a federal clearinghouse known as Gold Eagle was stood up for sharing AI cyber threat information between the government and private sector.

You can read the full letter here.

The post 100-plus companies call for ‘global surge’ in AI-powered cyber defense appeared first on CyberScoop.

Artificial intelligence has never been more important to the federal government.

Under the Trump administration, AI has been adopted rapidly across the private sector and federal agencies. Software developers now use large language models to generate much of their code. Frontier AI models are escaping testing sandboxes to hack live internet infrastructure. Foreign governments are conducting cyber and kinetic attacks targeting data centers and other AI-related infrastructure.

The AI industry’s lightning-fast evolution since 2022 and growing importance to U.S. economic and national security have prompted calls  for stronger federal oversight in order to better manage emerging threats.

A new report published Thursday from the nonprofit Americans for Responsible Innovation, shared exclusively with CyberScoop, calls for the federal government to declare key AI models, companies and its supporting industries as critical infrastructure. It also calls for naming the Cybersecurity and Infrastructure Security Agency as the lead agency managing cyberthreats for the sector.

The report defines the AI sector as organizations, facilities, technologies, and industries “whose primary purpose is the development, training, deployment, and operation of AI systems.” It includes frontier model designs, model weights, evaluation and alignment systems, datacenters and AI-specific hardware, semiconductor chips and the platforms and infrastructure used to deploy and serve AI models at scale.

“The AI sector already bears all the hallmarks of critical infrastructure,” wrote authors Terrence Kelly and Jessica Maksimov. “It is interwoven with public and private services, concentrated among a handful of foundation models, and increasingly interdependent with [critical infrastructure] sectors, meaning a single attack on the AI stack could cascade across multiple sectors at once.”

In an interview, Maksimov told CyberScoop that while there are other options, CISA makes the most sense to lead the sector’s cybersecurity efforts because of its statutory mission, experience managing eight other critical infrastructure sectors and background dealing with cybersecurity problems that cross different sectors and industries.  

“We want an agency that has coordination authority across all other departments, because we believe that AI will just be so prevalent across different infrastructure [impacting] finance, energy, government services, that’s already equipped to coordinate across the entire interagency and talk about infrastructure in that way,” she said.

The U.S. is particularly susceptible to AI supply chain disruptions because frontier AI companies and most of their computing resources are based in the country. As the Trump administration pushes broader adoption across government and the private sector, experts warn that a major disruption could have outsized economic consequences. 

The past year has offered a potential vision of that future, with Iranian drones attacking Amazon-owned datacenters and Ukrainian drones striking Russian e-commerce giant Wildberries, causing disruptions to critical internet services.

“I would say that because of the value that attackers would put on U.S. AI capabilities and systems, that the infrastructure that supports all those capabilities is very vulnerable,” to both physical and cyber attacks, Maksimov said.

There are currently 16 critical infrastructure sectors managed by the federal government, and the designation carries real weight in terms of how departments and agencies prioritize their limited resources.

Matt Hayden, a former assistant secretary of homeland security for cyber infrastructure risk and resilience, said designating a sector or industry as critical infrastructure means the government puts you in a special category where you’re “identified as being a component of a national critical function that the U.S. population, the economy, depend on.”

The designation unlocks a wide range of federal tools and resources, often free of charge, including operational continuity and incident response services, cybersecurity software, access to federal systems like Continuous Diagnostics and Mitigation (CDM), and bespoke, real-time threat intelligence.

Hayden said that the AI ecosystem described in the report captures many critical industries, and he believes that at the very least, frontier models will one day be covered as critical infrastructure, whether through a new designated sector or existing ones, like the IT and telecommunications sector.

But he noted that other sectors, such as space or cloud computing, have similarly argued for a critical infrastructure designation. He also predicted that any effort to formalize a federal lead for AI security would result in a bureaucratic turf war. Under the Trump administration, the Departments of Commerce and Treasury have played more prominent roles in shaping policy and regulation around AI systems.

“We have fought those battles in the policy circus for trying to get space-based critical infrastructure carved out, and it’s as complicated as trying to find an owner,” said Hayden, now a vice president at General Dynamics Information Technology. “Everyone in the government has to agree that that [agency] is the primary, and as a result it’s very difficult to get those documents across the finish line.”

Hayden also said that new programs like ANCHOR-CI allow CISA to quickly convene ad-hoc stakeholder meetings to address emerging cyber threats. It also gives the CISA director authority to add individual companies to existing critical infrastructure sectors.

Bob Kolasky, former director of the National Risk Management Center at CISA, told CyberScoop that he believes companies like OpenAI and Anthropic, as well as data center operators, will eventually be designated as critical infrastructure. He said it’s still an open question whether ANCHOR-CI, which was rolled out by DHS in July, will be an improvement over the existing processes scrapped by the Trump administration last year.

“Every sector is going to rely on artificial intelligence and making more resilient the sectors themselves and understanding how they use AI and the dependencies” is still going to be an important task, said Kolasky, now senior vice president of critical infrastructure at Exiger.

And while CISA is well-positioned as a potential lead, Kokasky said the AI sector is likely to bring its own unique set of challenges and coordination issues.

“If you just sort of layer on another sector and say ‘function like the other 16 sectors,’ right now, those 16 sectors are all over the place in terms of how they’re functioning,” said Kolasky. 

The post The push to designate AI as the next critical infrastructure sector appeared first on CyberScoop.

A company that runs AI testing labs involved in a series of breaches carried out by Anthropic and OpenAI’s cyber-focused models said the incident happened in part because they “unintentionally” provided the models with access to the internet.

In a blog posted Friday, Irregular said it partners with frontier labs to “stress test” AI models for security risks before they’re deployed, including simulations of their vulnerability research capabilities.

Those tests included non-public models from Anthropic and OpenAI. Separate disclosures from the companies confirmed these included Mythos 5, Claude Opus and GPT-5.6 Sol.

“As part of our review, we identified that a few interactions with our evaluation environments, in which internet access was unintentionally made available, led some models to take offensive security actions in the real world,” Irregular said in an unsigned blog.

In the incidents involving Anthropic’s models, testers initially believed the models were using a fictional company in its hacking simulations, only to learn later that the name unintentionally matched the domain of a real company. That led the model to “consider taking offensive action against” the real company in multiple training runs.

Irregular said this was due to “human oversight,” and that they typically do a background check to ensure they’re not using a real company for cybersecurity tests. The instructions provided to the model also included the target’s name and “internal addresses within our network” where the model could go to find the simulated environment.

While Mythos performed as intended in the majority of the thousands of tests, a small number of cases showed the model couldn’t distinguish between the fake and real domains and executed actual attacks on internet infrastructure, including exploiting vulnerabilities, extracting credentials and accessing a production database. In one instance, it even began targeting a site with a similar name when credentials for it were discovered online.

“Mainly, models believed they were in simulated environments, when they in fact took action in the real world,” the blog continued. “We are putting in place new and robust protocols to ensure setup issues do not occur while meeting the constraints of the testing process.”

The company said it plans to release a larger whitepaper breaking down the incidents and update their best practices for evaluation setups in the future.

While the companies have drawn criticism from some in the cybersecurity community for failing to securely design their sandboxes for testing, experts have said AI models are known to grind away on fulfilling a command  until they can find a workaround. Additionally, Irregular said granting some level of internet access to models is necessary to fully test out their cybersecurity capabilities.

“Controlled internet access, while it may allow models to exceed containment boundaries, is at times critical for realistic evaluations; without it, threat scenarios lose fidelity, undercutting the purpose of the challenge to reduce post-release risk of models being misused by attackers – as attackers in the real world do rely on the internet,” the company wrote.

According to the blog, Irregular has since “remediated” the “issues that led to these interactions,” though few details are provided.

However, the researchers say the engagement revealed critical gaps in their security practices. 

They plan to improve documentation of evaluation setups, deploy better log monitoring tools capable of tracking “the extreme amount of data generated by the traffic,” revise their threat models to account for rogue AI behavior, and establish faster information sharing between stakeholders.

“Looking further down the line, models will only get stronger. While in this case we believe that better implementation of existing safeguards could prevent most incidents of this kind, as models become stronger, this may not be the case,” Irregular wrote. “We therefore believe this opportunity should be leveraged by us and the community to be proactive and establish forward-looking protocols and [research and development] efforts.”

The post Irregular says ‘human oversight’ responsible for AI sandbox escape incidents appeared first on CyberScoop.

As the White House and federal agencies grapple with frontier AI models and their hacking capabilities, researchers are warning that the industry’s “middle class” of smaller models may end up posing a greater threat over the long term.

Research from XBOW this week shows that a growing class of both proprietary and open-source models are becoming strategically important in the offensive security ecosystem. Models like Z.ai’s  open-weight GLM-5.2, xAI’s Grok 4.5, Anthropic’s Opus 4.7, Meta’s Muse Spark 1.1, still perform very strongly at many hacking and exploitation tasks that worry policymakers.

“It’s not even that the open-source variants or…not quite frontline competitors are catching up [to frontier models] as such,” said Albert Ziegler, head of AI at XBOW. It’s that they are crossing a certain threshold, which means that suddenly they are providing net value at a cheaper price.”

That wasn’t necessarily the case as recently as six months ago, when testing on mid-tier class models showed they struggled to complete “moderately complex” agentic tasks. Today’s middle class largely can. Their relative cheapness means users can spend many times more resources—running them repeatedly—to solve the same challenges. 

“Because these models are cheaper, it’s okay to give them more time, and they come from behind and leapfrog the big frontier model,” said Ziegler. “Now, that didn’t work half a year ago because…if you wanted to run some open-source model on a complex task in an agentic way…on a long horizon, then it would just get lost.”

GPT 5.5, now considered a near-frontier model, delivered one of the best performances on exploitation benchmarks that XBOW has recorded to date.

The jump between OpenAI’s GPT 5 and 5.5 “represented one of the clearest 2026 leaps in autonomous web application testing,” according to the report. It saw marked improvements over previous middle-class models in exploiting both “white box” and “black box” scenarios, or with and without access to the underlying victim source code. It also missed fewer vulnerabilities, with a “miss rate,” or failure to spot a vulnerability, of 10%, while GPT 5’s rate was four times larger, 40%.

The emergence of GPT 5.5 changed “the practical baseline for what frontier models can do in offensive workflows,” the XBOW report said.

But the performance leap goes deeper than that. GPT 5.5 performed higher in tests without source code access, while GPT 5 heavily leaned on source code. 

“That last result is significant: working without the code, as an attacker would, GPT-5.5 beat a prior version that could read it,” the XBOW report said. “What translated into findings was the ability to reach and prove a vulnerability against the running system, not to infer it from a pattern in the source.”

XBOW’s testing found that source code access was not as important to these models’ success as other factors, like live interaction with the actual website or software being exploited.

Frontier models like Mythos and GPT 5.6 are indeed more capable on individual cybersecurity tasks, but they can also come with exponentially higher token costs.

New research this week from Anthropic tested two models – Mythos Preview, which is used in Project Glasswing, and Opus 4.8 – to learn how quickly multi-agent swarms could find vulnerabilities in 15 open-source software projects when they coordinate and share information.

While a team of agents working individually and assigned to core directories found 21 vulnerabilities, the coordinating agent swarm found 266. But both tests had to burn through millions of tokens – 6.5 million and 27 million – to get there. Beyond the difficulties with getting access to frontier models, few individuals or organizations have the budget to underwrite that kind of research.

The way these systems coordinate can differ significantly from how humans work together.

Another experiment tested agents’ ability to coordinate on the development of a fantasy-themed video game. Earlier models, models like Opus 4.6, failed to properly coordinate and produced “bad” results, while later models like Mythos and Opus 4.8 were able to achieve better results but did so by hardly coordinating at all on tasks.

“The lack of coordination shown by agents in the fantasy game…in which they siloed themselves and largely failed to merge their work—roughly mirrors some ways in which humans can fail to coordinate,” the Anthropic blog stated.

Further, agents are more homogeneous than humans and “often act the same in situations where different people might take a much more diverse range of actions.”

XBOW also tested Mythos Preview, finding that it showed “exceptional” source-code reasoning and reverse engineering abilities, particularly with source code access. Like other models, losing live-site access had a big impact on its performance, and while Mythos is excellent at finding vulnerabilities, it’s less effective at exploiting them.

]Ziegler said the recent incidents at companies like OpenAI, Anthropic, Meta and others where frontier models escaped sandboxes and hacked into project-adjacent parts of the internet should rightfully alarm lawmakers, and demonstrate  the upper-tier capabilities of large language models.

Like most industries, cybersecurity favors cheap, high-performing tools over expensive ones. The widely adopted tools that have the most impact tend to be affordable and effective, not luxury products. 

And while these models still require human management to be wielded responsibly by law-abiding organizations, that cost tradeoff can look more attractive to malicious hackers, who tend not to care about collateral damage caused by their agents.

“Purely from an attacker’s perspective, I think we already are [there],” Ziegler said.

The post AI’s ‘middle class’ has gotten dramatically better at hacking appeared first on CyberScoop.

As security operations teams now use large language models (LLMs) and autonomous AI agents into their daily work, a new frontier is emerging: attackers deliberately manipulating AI agents. Prompt injection attacks—where an attacker hides malicious instructions that cause an AI agent to ignore its safety rules—pose a serious risk to enterprises. These attacks continue to grow in size and scale.  

Snyk’s security audit of the Agent Skills ecosystem, which includes Anthropic’s Claude, Vercel, and others, that 36% of all skills contained at least one critical-level security issue, including malware distribution, prompt injection attacks, and exposed secrets.

In June, researchers at Mozilla tested a prompt injection attack on Claude using indirect prompt injection—a technique that embeds malicious instructions in external content the AI agent processes. In this proof-of-concept, attackers took over developers’ systems by hiding indirect prompts in normal-looking repositories. When Claude Code executed them, the agent spawned a reverse shell.

AI agents often connect to more sensitive data than human employees do., A successful prompt injection can lead to catastrophic data loss or unauthorized system actions. Defending against prompt injection attacks requires multiple layers of protection. Security teams must monitor agent behavior for anomalies and prepare for agent containment, forensic preservation, and system remediation. Because AI agents execute tasks at machine speed, human responses must be able to match that pace.

The architecture of trust: Protocols and no “black box”

AI-native workflows need governed access rather than “black-box” autonomy. Modern governance frameworks use standardized protocols like the Model Context Protocol (MCP) to provide secure communication between AI clients and data sources. Visibility and transparency in agentic AI workflows matter, especially in cybersecurity. Autonomous agents perform complex tool executions and use independent logic, so they must show how they reached their decisions to meet regulatory requirements. Agents without transparency post serious risks: obscured reasoning can trigger unpredictable tool interactions, bypass governance controls, and create uncontrolled defensive gaps.

Implementing these protocols matters:

• Bounded Tenant Awareness: In a stable agentic AI architecture, multi-tenancy scales well. But if an AI tenant misbehaves, the entire system can fail. Bounded tenant awareness isolates any misbehaving AI agent to prevent cross-tenant contamination or data leakage.
• Strict Access Controls: By controlling connections to the platform, organizations can stop “ignore previous instructions” style bypasses. Maintain tight control over what the AI can see and do within a workflow.
• Standardized Telemetry: All telemetry must remain consistent and audit-ready. Even if an AI interaction is attempts to break rules, the underlying data movement gets tracked against established frameworks like MITRE ATT&CK and NIST.

Detecting the aftermath: UEBA and NDR as safeguards

A robust, unified SecOps platform can detect anomalous behavior even after prompt injection tricks an AI agent. Prompt injections often serve to steal credentials theft or extract data. When detected it’s important to act quickly. In agentic AI systems, misbehavior can escalate privileges, manipulate memory layers, create unauthorized identities, or alter shared reasoning components. Containment must be automatic and enforced at identity, authentication, and authorization layers.

These safeguards include:

• User and Entity Behavioral Analytics (UEBA): Identity-focused correlation and behavioral baselines to identify anomalous user activity or privilege escalation. If a compromised AI agent acts outside of its normal operational parameters, UEBA flags it in real-time and alerts a human security analyst.
• Network Detection and Response (NDR): Combining network traffic analytics with endpoint and cloud telemetry, NDR can identify data exfiltration or policy violations from a successful prompt injection.
• Multi-Layer AI Filtering: AI filters reduce raw alerts into high-fidelity incidents, cutting noise by up to 90%. This keeps the signals of an AI-driven attack from disappearing in a busy SOC.

Humans remain the strongest defense against AI agent social engineering. The human security analyst is still the one who makes the final decision. While AI handles triage and correlation, humans retain final control over response actions.

Moving beyond reactive guardrails

The traditional SOC model was never designed to handle machine-speed, AI-driven attacks. A human-augmented autonomous SOC approach moves from reactive alert handling to a proactive, verdict-first model. By combining a transparent, governed AI access with robust UEBA and NDR, organizations keep the SOC secure, transparent, and resilient as social engineering methods target machines.

The post Why transparent AI agents matter more than you think appeared first on CyberScoop.

As AI-generated code continues to be injected into all corners of the internet, concerns have risen about an expanding attack surface for malicious hackers to exploit.

Some have argued that the enhanced cybersecurity capabilities of large language models could serve as a check, finding and fixing vulnerabilities nearly as fast as they’re created.

But new research that tested the patching capabilities of two popular commercial models, OpenAI’s ChatGPT 5.5 and Anthropic’s Claude Opus 4.8, found that generative AI is more likely to create an exploitable patch or introduce entirely new bugs than close off a vulnerability.

Researchers at 1Password tested the models ability to patch six “high-impact, high-complexity” CVEs, including the “Copy Fail” vulnerability, a kernel flaw that can give an attacker root access to Linux cloud environments. The overall success rate (or fully patching the vulnerability without introducing new problems), was less than a coin flip at 47%.

“Our research findings show that, in aggregate across a variety of scenarios, both Claude and ChatGPT had a low rate of successful patch generation, which we define as full remediation of all known exploit paths with no erroneous changes to application behavior,” wrote Keith Hoodlet, Axel Mierczuk and Spencer Michaels.

“The models often addressed only a subset of vulnerable code paths, added fragile guard code that satisfied tests while failing to address the vulnerability’s root cause, and sometimes introduced subtle changes in the application’s behavior while patching the immediate vulnerability,” the authors continued.

The research suggests that largely autonomous vulnerability-discovery and patching may not yet be effective in fixing the explosion of vulnerable code that is being created in the AI era.

Other private sector research has pointed to a similar problem. A report this year from Veracode found that while LLMs have made “enormous strides” in crafting workable code, “security is a different story.” Testing across a range of frontier models found the average security “pass rate” for AI generated code is around 56%. Newer models like GPT 5.5 push closer to 70%, while more than half sit between 50-53%.

Veracode tested 100 different models and while there was variability, in general a small number of models were showing progress on security patching while the rest have experienced “stagnation.” Similar to the 1Password research, in 44% of Veracode tests the models introduced a detectable OWASP Top 10 vulnerability into the codebase.

An important caveat: neither report tested newer models, like Anthropic’s Mythos or OpenAI’s GPT-5.6-Sol, that frontier companies tout as having significantly higher cybersecurity capabilities.

Those advanced models can identify and fix vulnerable code. Anthropic and OpenAI are distributing them to key industries through Project Glasswing and Daybreak before foreign or open-source alternatives can compete.

Tim Jarret, vice president of product at Veracode, told CyberScoop that AI tools are still subject to a range of limitations that can make them unreliable for cybersecurity patching without knowledgeable humans in the loop.

While some vulnerabilities – like SQL injections – can be easily patched through automation, other bugs like cross-site scripting, can be exploitable in several different ways and require either a human touch, additional context or both to fully close off. Additionally, models can slowly lose context from prior sessions over time, affecting their ability to complete tasks correctly and raising the possibility they’ll hallucinate to fill in the missing gaps.

“I think we would say, at this point, that Iits premature to treat those as anything other than another code change to the code base that needs to be reviewed and accepted by the team, as opposed to letting the agent merge the code freely,” said Jarrett.

However, he acknowledged that may not be possible in a world where AI agents are generating exponentially more code for human defenders to review. Some kind of automated code review will be necessary – preferably not by the same automation tool that produced the code. The ultimate goal is the same as it has always been in security: “trust but verify.”

“Ninety percent of the time, the human check might just be ‘did the cross check look good?’ Do we have a thumbs up?’” Jarrett said. “In those cases where there’s still something wrong, that’s where you focus your attention a little bit more.”

The post More than half of AI-generated patches are broken appeared first on CyberScoop.

The UK’s AI Security Institute said that their AI research system took “unsanctioned” actions over the internet to engage in “sustained, potentially harmful activity directed at real people.”

According to a blog post and technical report released Tuesday, the organization noticed “unusual data transfers” leaving their research systems through the pseudo-anonymous Tor network on July 28. The models, which included Anthropic’s Mythos 5 and OpenAI’s GPT-5.6-Sol models, were being tested for their cybersecurity capabilities.

Like the OpenAI and Anthropic incidents before it, the offending agents were attempting to solve a challenge. On 10 of 122 runs taken by two models, they took a combined 19 malicious actions, including attempting to insert malicious code into a real open-source software project and creating fake online identities to contact human maintainers and press them to approve it.

It also inserted malicious prompt injection instructions in places “where it reasoned that other automated AI systems might pick them up and execute them.” Sometimes multiple agents would collaborate on a task, with one model leaving public messages or notes for the other on GitHub.

However, AISI said their incident differs in one critical way.

“Importantly, this was not a case of a model escaping its secure test environment, or ‘sandbox,’” the institute wrote in an unsigned blog. “As was standard in our cyber testing, we had intentionally permitted internet access, and model-provider cyber classifiers were deliberately disabled – conditions that do not reflect how frontier models are made available to the public.”

The institute said that while their own design and evaluation choices “to some degree” enabled the malicious behavior, the models still displayed “novel, potentially deceptive behaviors” and were executed “to an extent and severity that we did not anticipate.”

The same day, OpenAI acknowledged AISI and another third-party cybersecurity tester,  private firm Irregular, had recently reported incidents where their models exceeded “beyond their intended testing boundaries.”

That included GPT-5.6-Sol reusing GitHub tokens left by another agent to attempt to recover accounts and to access a DNS server that contained malicious payloads meant to exploit a software vulnerability.

The setup did not work, there is no evidence any real resolver queried it, and the infrastructure was removed when the evaluation ended,” the company wrote in an unsigned blog post Tuesday.

OpenAI said it would review its own third-party testing procedures to focus on higher risk evaluations, assess requests by third-parties to enable internet access, stop conditions and other features for their models.

The incident with Irregular occurred on July 29 during a Capture-the-Flag cybersecurity evaluation of OpenAI’s models. Due to a “misconfiguration” the models were allowed to access the public internet, where they encountered and exploited a real domain, mistakenly believing it was still in a test environment.

OpenAI said an investigation by Irregular is ongoing, but also found that the models had found and used credentials for the site at one point. The blog also references other additional potential cybersecurity incidents.

“Irregular has informed us that all of the issues identified pertaining to the incident are no longer active and relevant safeguards were added to the testing environment,” the blog said. “Irregular has also communicated about related incidents involving other labs from the same testing environment.”

CyberScoop has reached out to Irregular for comment.

The incidents were made public the same day that the White House met with Anthropic, Open AI and other frontier AI companies to preview a new framework for evaluating models before they’re released publicly. Some media outlets have reported that after an executive order, export controls and other actions, the administration does not plan to make the new framework public.

The post AISI, OpenAI report more ‘unsanctioned’ model hacks appeared first on CyberScoop.