This is a fascinating explotation of how LLMs fall for prompt injection attacks. It turns out that they learn to recognize the style of text in different role/instruction blocks, and not just the tags.

Their conclusion:

Role tags were a formatting trick that became the security architecture and the cognitive scaffolding of modern LLMs. We’ve shown that this architecture doesn’t survive into the model’s actual representations, and that such role confusion is linked to prompt injection.

Unless LLMs achieve genuine role perception, we think injection defense will remain a perpetual whack-a-mole game. And the continuous nature of role boundaries opens the threat of injections designed to subtly shift LLM states through seemingly innocuous text, legally and at scale...

At least one malware developer is adding text about nuclear and biological weapons to their spyware, in an effort to stop automatic AI analysis.

Details:

The _index.js payload begins with a large JavaScript block comment containing fake system instructions and policy-triggering content. Because it is inside a comment, it does not affect JavaScript execution. The runtime skips it. The real malware begins after the comment with a try{eval(…)} wrapper around a large character-code array and a ROT-style substitution function.

This header appears designed for AI-mediated analysis, not for Node, Bun, or Python. It attempts to derail scanners or analyst copilots that feed the beginning of a file to a language model without clearly isolating the content as untrusted data. In weak pipelines, this can cause refusal behavior, prompt confusion, context pollution, or premature classification before the scanner reaches the actual malware...

Fable 5 is the supposed safe version of Anthropic’s Mythos Preview, with guardrails to ensure that it can’t be used to create cyberattacks.

Well, that restriction was bypassed within days.

I haven’t thought about the privacy issues surrounding professional athletes and wearables.

Wearables present serious privacy issues for “Average Joe” consumers, who are entrusting tech companies to safely store and protect their biometric data. Imagine the stakes for a professional athlete, whose entire livelihood could be affected by a single biometric data point. To give one of many realistic hypotheticals: a basketball player has a terrible game, and the coach wonders if they showed up to the gym hungover. The coach has access to the player’s wearable data, and checks to see when they went to sleep, as well as what their heart rate looked like during the night. Should the player have been out partying before a game? No. Should the coach be able to surveil them? Definitely not...

Dolphins, sharks, turtles, and human workers are all victims of unregulated squid fishing fleets.

Another news article.

As usual, you can also use this squid post to talk about the security stories in the news that I haven’t covered.

Blog moderation policy.

On June 9th, Anthropic released its Fable generative AI model. Three days later, the US government classified it as a dangerous munition, and used its export-control authority to prohibit any foreign nationals from accessing it. Unable to differentiate between Americans and foreigners, the company shut off access for everyone.

The government’s actions won’t help. The problem isn’t any one particular model; it’s the general trend of increasing AI capabilities. And any real solution requires the sort of collective action that just isn’t possible right now...

At least one malware developer is adding text about nuclear and biological weapons to their spyware, in an effort to stop automatic AI analysis.

Details:

The _index.js payload begins with a large JavaScript block comment containing fake system instructions and policy-triggering content. Because it is inside a comment, it does not affect JavaScript execution. The runtime skips it. The real malware begins after the comment with a try{eval(…)} wrapper around a large character-code array and a ROT-style substitution function.

This header appears designed for AI-mediated analysis, not for Node, Bun, or Python. It attempts to derail scanners or analyst copilots that feed the beginning of a file to a language model without clearly isolating the content as untrusted data. In weak pipelines, this can cause refusal behavior, prompt confusion, context pollution, or premature classification before the scanner reaches the actual malware...

On 14 April, the Trump administration quietly acknowledged the widespread use of AI to automate government processes. The office of management and budget (OMB) disclosed a staggering 3,611 active or planned use cases for AI across the federal government. The list has ballooned by 70% from the one published in the final year of the Biden administration, and includes many disturbing-seeming plans to hand over sensitive governmental functions to AI.

Scanning this list, many readers may find many causes for alarm. It represents a transfer of decision processes from human to machine on a massive scale over matters of individual freedom, public health and well-being, nuclear reactor safety and more...

There are over a dozen cases around the country where police officers are using the Flock surveillance camera system to obsessively and illegally stalk people.

Alternate link.

A proposed FCC rule would kill burner phones: phones whose accounts are not attached to a particular person.

The FCC plans to do this by legally forcing the country’s telecoms to store a wealth of personal information about essentially all phone customers, including a government issued identification number and their physical address, alarming privacy advocates and civil rights activists who compare the measures to those from authoritarian countries where it can be difficult to buy a mobile phone plan without giving up your identity.

The proposed change would drastically shake up how people obtain phone plans in the U.S., and have all sorts of privacy and cybersecurity knock-on effects. The FCC is proposing the data collection partly as a way to combat scammers, with telecoms being required to collect other information on business and foreign customers like the intended use case of their bulk phone plan purchase and their IP address. But the changes would mean telecoms collect data on all new and renewing customers, and the FCC provides a long list of other things that the collected data could help authorities with...

This is a current list of where and when I am scheduled to speak:

• I’m giving a keynote at Cybernation 2026 in Berlin, Germany, on June 24, 2026.
• I’m speaking at the Potsdam Conference on National Cybersecurity at the Hasso Plattner Institut in Potsdam, Germany. The event runs June 24–25, 2026, and my talk will be the evening of June 24.
• I’m participating in a panel discussion at the Austrian Institute for International Affairs in Vienna on Thursday, June 25, 2026.
• I’m speaking at the Digital Humanism Conference in Vienna, Austria, on Friday, June 26, 2026...

This fluid pump was inspired by the way squids propel themselves through the water.

As usual, you can also use this squid post to talk about the security stories in the news that I haven’t covered.

Blog moderation policy.

Let no one accuse Bernie Sanders of ducking the big questions. Writing in the New York Times last week, the senator asked: “Will the future of humanity be determined by a handful of billionaires who have promoted and developed AI, with virtually no democratic input, who stand to become even richer and more powerful than they are today?”

We agree entirely that this is one of the most potent questions facing global democracy today. Our book, Rewiring Democracy, surveys the emerging uses for and impacts of AI in democracy around the world and reaches the same conclusion: that the most urgent risk posed by AI is the ...

The surveillance company Leonardo wants more data:

A surveillance company plans to add sensors to automatic license plate readers (ALPRs) that would mean the devices, as well as capture the license plate of passing vehicles, would also sweep up unique identifiers of mobile phones, wearables, and other Bluetooth-enabled devices in those cars, potentially letting law enforcement identify specific drivers or passengers.

The technology, called SignalTrace, would turn ALPR cameras from devices focused on tracking cars to ones that can more readily track the location of particular people. ALPR cameras have become a commonly deployed technology all across the U.S.; SignalTrace would make some of those cameras capable of collecting much more data...

WhatsApp has caught the NSO Group phishing its users, in violation of a court order.

This is interesting:

The U.S. military has likely been quietly broadcasting codes for its global encryption network using public GPS for nearly 20 years, turning each satellite into a hidden “numbers station,” according to Steven Murdoch…

That means every device that uses GPS has been receiving hidden government information for years, and nobody outside the military knew it until now.

[…]

Murdoch discovered that this particular sentinel was transmitted by all 31 operational satellites within a window of a few hours on May 26, 2011, potentially heralding the activation of a new operational system. He confirmed that this timeline coincided with the rollout of the military’s Over-the-Air Distribution (OTAD) and the Over-the-Air Rekeying (OTAR) by cross-referencing declassified documents, including a 2015 presentation about the dates of the operation...

If you’re a user—owner?—of this cryptocurrency, this is important:

On May 29, the security researcher Taylor Hornby found a critical vulnerability in Zcash Orchard privacy pool using Claude Opus 4.8. The Zcash team hired Hornby specifically to look for this kind of issue. He found one fast enough to be embarrassing.

The Orchard pool is the newest and most advanced shielded transaction system in the cryptocurrency Zcash. Introduced in 2022, it allows users to send and receive ZEC while keeping transaction details private. It uses zero-knowledge proofs to validate transactions without revealing amounts or participants. The bug: a specific check that was supposed to validate transaction inputs wasn’t actually enforcing the rules it appeared to enforce. An attacker could have exploited the flaw to feed false inputs into that check and generate ZEC from nothing, with the zero-knowledge proof system blessing the fraudulent transaction as valid...

In April, Anthropic initated Project Glasswing. The idea was to let companies use their new model to find and fix vulnerabilities in their own software. It was a fantastic PR move, and so many press outlets have uncritically parroted Anthropic’s claims that it’s now common wisdom that Mythos is better at finding software vulnerabilities than other models. Which is just not true.

In any case, Anthropic has published a Project Glasswing status report. It’s finding a lot of vulnerabilities in software—yay! Some of them are even dangerous. But almost none of them has been patched. It’s ...

Researchers have prototyped an AI-powered internet worm.

The coolest thing about the prototype is that it carries its own LLM with it, and runs it on computers that have been broken into.

This is the closest to John Brunner’s original 1975 conception of a computer worm that I’ve seen.

Hackers are convincing Meta’s AI support chatbot to let them take over other peoples’ accounts:

A video posted on X showed the step-by-step process to hack someone’s Instagram account. The hacker allegedly used a VPN to spoof the targets’ presumed location to avoid triggering Instagram’s automated account protections. Then, the hacker opened a chat with Meta AI Support Assistant and asked the bot to add a new email address to the target’s account. The chatbot can be seen sending a verification code to the email address provided by the hacker; the hacker then shares the verification code with the chatbot, which prompts the chatbot to show a button to “Reset Password.” The hacker enters a new password and takes over the victim’s account...