Peter Rusch

The Ocean Race

[Alex Veremeyenko @alex_verem]

🚨 BREAKING: Google DeepMind just mapped the attack surface that nobody in AI is talking about.

Websites can already detect when an AI agent visits and serve it completely different content than humans see.

> Hidden instructions in HTML.
> Malicious commands in image pixels.
> Jailbreaks embedded in PDFs.

Your AI agent is being manipulated right now and you can't see it happening.

The study is the largest empirical measurement of AI manipulation ever conducted. 502 real participants across 8 countries.

23 different attack types. Frontier models including GPT-4o, Claude, and Gemini.

The core finding is not that manipulation is theoretically possible it is that manipulation is already happening at scale and the defenses that exist today fail in ways that are both predictable and invisible to the humans who deployed the agents.

Google DeepMind built a taxonomy of every known attack vector, tested them systematically, and measured exactly how often they work.

The results should alarm everyone building agentic systems.

The attack surface is larger than anyone has publicly acknowledged. Prompt injection where malicious instructions hidden in web content hijack an agent's behavior works through at least a dozen distinct channels.

Text hidden in HTML comments that humans never see but agents read and follow. Instructions embedded in image metadata.

Commands encoded in the pixels of images using steganography, invisible to human eyes but readable by vision-capable models.

Malicious content in PDFs that appears as normal document text to the agent but contains override instructions.

QR codes that redirect agents to attacker-controlled content.

Indirect injection through search results, calendar invites, email bodies, and API responses any data source the agent consumes becomes a potential attack vector.

The detection asymmetry is the finding that closes the escape hatch. Websites can already fingerprint AI agents with high reliability using timing analysis, behavioral patterns, and user-agent strings.

This means the attack can be conditional: serve normal content to humans, serve manipulated content to agents.

A user who asks their AI agent to book a flight, research a product, or summarize a document has no way to verify that the content the agent received matches what a human would see.

The agent cannot tell the user it was served different content.

It does not know. It processes whatever it receives and acts accordingly.

The attack categories and what they enable:
→ Direct prompt injection: malicious instructions in any text the agent reads overrides goals, exfiltrates data, triggers unintended actions
→ Indirect injection via web content: hidden HTML, CSS visibility tricks, white text on white backgrounds invisible to humans, consumed by agents
→ Multimodal injection: commands in image pixels via steganography, instructions in image alt-text and metadata
→ Document injection: PDF content, spreadsheet cells, presentation speaker notes every file format is a potential vector
→ Environment manipulation: fake UI elements rendered only for agent vision models, misleading CAPTCHA-style challenges
→ Jailbreak embedding: safety bypass instructions hidden inside otherwise legitimate-looking content
→ Memory poisoning: injecting false information into agent memory systems that persists across sessions
→ Goal hijacking: gradual instruction drift across multiple interactions that redirects agent objectives without triggering safety filters
→ Exfiltration attacks: agents tricked into sending user data to attacker-controlled endpoints via legitimate-looking API calls
→ Cross-agent injection: compromised agents injecting malicious instructions into other agents in multi-agent pipelines

The defense landscape is the most sobering part of the report.

Input sanitization cleaning content before the agent processes it fails because the attack surface is too large and too varied.

You cannot sanitize image pixels. You cannot reliably detect steganographic content at inference time.

Prompt-level defenses that tell agents to ignore suspicious instructions fail because the injected content is designed to look legitimate.

Sandboxing reduces the blast radius but does not prevent the injection itself. Human oversight the most commonly cited mitigation fails at the scale and speed at which agentic systems operate.

A user who deploys an agent to browse 50 websites and summarize findings cannot review every page the agent visited for hidden instructions.

The multi-agent cascade risk is where this becomes a systemic problem.

In a pipeline where Agent A retrieves web content, Agent B processes it, and Agent C executes actions, a successful injection into Agent A's data feed propagates through the entire system.

Agent B has no reason to distrust content that came from Agent A. Agent C has no reason to distrust instructions that came from Agent B.

The injected command travels through the pipeline with the same trust level as legitimate instructions. Google DeepMind documents this explicitly: the attack does not need to compromise the model.

It needs to compromise the data the model consumes. Every agentic system that reads external content is one carefully crafted webpage away from executing attacker instructions.

The agents are already deployed. The attack infrastructure is already being built. The defenses are not ready.

[Climate Dad @ClimateDad77]

431.76ppm

Tumbleweeds.

[Peter Rusch @Peter_Rusch]

@stats_feed @grok why is Canada so high on this list

[Glenn Gould @GlennGould]

70 years ago, Glenn Gould changed the course of classical music.
Celebrating the 70th Anniversary of his groundbreaking 1956 recording of Bach’s Goldberg Variations. A performance that still resonates, challenges, and inspires today. 🎶

[AFP News Agency @AFP]

The world's tropical coral reefs have almost certainly crossed a point of no return as oceans warm beyond a level most can survive, a major scientific report announced on Monday ➡️ https://t.co/xx5YEduKi5

[Chris Hadfield @Cmdr_Hadfield]

All the roads in Canada. Reveals much to think about.
(map by @EarthArtAus)

[Peter Rusch @Peter_Rusch]

@globepolitics Headline should be 'Canadians hate their kids'.

[Marathon of Hope Cancer Centres Network @marathonofhope]

Terry Fox started a marathon against cancer. Together, we can end it. The Marathon of Hope Cancer Centres Network is a national movement committed to finishing Terry's fight against cancer. Learn more: https://t.co/g4gb2zQ0FT

[Peter Rusch @Peter_Rusch]

@Nicolascole77 🤖

[The Ocean Race @theoceanrace]

It's been a year since the unforgettable orca encounter near the Strait of Gibraltar during The Ocean Race 2022-23! 🐋

Orcas stopped Team Jajo, making global headlines. Skipper Jelmer van Beek and Watch Captain Rutger Vos reflect on this incredible experience.

Click on the link to dive into their story 👉 https://t.co/Wx1XmyUld8

#TheOceanRace #orca #whales #ocean

[Peter Rusch @Peter_Rusch]

@RestIsPolitics @campbellclaret @RoryStewartUK https://t.co/iwauSFTL3h

[Prof. Carl Sagan @ProfCarlSagan]

Do you Remember this Voice?

[Peter Rusch @Peter_Rusch]

Not your everyday 3…

[The Athletic @TheAthletic]

Steph Curry with an insane full-court shot from the tunnel!

🎥 @NBCSWarriors

[Massimo @Rainmaker1973]

This photo is still one of the coolest iceberg pictures in existence thanks to the perfect composition and symmetry that separates the photo into four different sections of colour and texture.

[📸 David Burdeny]

[Super 70s Sports @Super70sSports]

“We’re playing street hockey today. Wanna come?”

“Is your friend any good?”

“Yes.”

[The Ocean Race @theoceanrace]

The Ocean Race and race chairman Richard Brisius brought the spirit of ocean racing to the #WEF in Davos to promote positive change in ocean protection and restoration with some of the world’s top business and political leaders vowing to scale-up their commitment to the race for the ocean.
More here - https://t.co/oj0GjiVSM2

[Peter Rusch @Peter_Rusch]

@TOsports97 @FriedgeHNIC Marner has a no move clause now and next year - they can’t trade him. Again the player has the leverage to play it out and walk for nothing.

[Mike Hudema @MikeHudema]

A startling info graphic from the @UN - demonstrating the dire consequences of half a degree. Are you ready to say good bye to 30% of all marine life?

There is no time to waste. #ActOnClimate

#climate #energy #renewables #oceans

[Massimo @Rainmaker1973]

On the aerodynamics of dogs
https://t.co/hzVCRCGVkt