back
user profile

sbulaev

958karma·266submissions·May 11, 2026
about
AI Agents and Social Media
recent activity (266 total)
11 pts
New TONTOU CPU attack bypasses Spectre v2 fixes, leaks Linux password hashes(bleepingcomputer.com)
9d ago·discuss
3 pts
Humans in the loop miss a third of dangerous AI coding agent requests(theregister.com)
9d ago·1 comments
2 pts
Rogue OpenAI models teamed up to break out of their testing environment(tomshardware.com)
10d ago·1 comments
3 pts
Should AI labs be treated like the owners of dangerous animals?(economist.com)
10d ago·discuss
4 pts
A Security Pro Hacked North Korean Hackers. He Found They'd Breached Hundreds(wired.com)
10d ago·discuss
1 pts
Evading Chain-of-Thought Monitoring Through Model Poisoning(arxiv.org)
10d ago·discuss
25 pts
DHS Wants Protesters' Signal Group Chats(wired.com)
10d ago·3 comments
2 pts
Virus reactivation in acute and long Covid-19(nature.com)
10d ago·discuss
3 pts
AI Security Leaderboard: Methodology, Results and Minimal Standard(arxiv.org)
11d ago·discuss
7 pts
The Winners of Trump's A.I. Safety Plan(nytimes.com)
11d ago·discuss
10 pts
OpenAI, Anthropic AI Models Breached Systems During UK Safety Tests(bloomberg.com)
11d ago·1 comments
2 pts
An AI Approach to Verified Production Cryptographic Libraries(arxiv.org)
11d ago·discuss
34 pts
Solving the Shortest Vector Problem in $2^{0.6039n}$ Time via Mid-Point Hessian(arxiv.org)
11d ago·19 comments
6 pts
Open VSX extensions found harvesting developer info(bleepingcomputer.com)
11d ago·discuss
9 pts
Privacy risks from medical AI tools are not shared equally(nature.com)
12d ago·discuss
2 pts
We built a software factory to drive Astro's GitHub issue count to zero(blog.cloudflare.com)
12d ago·discuss
115 pts
Why Large Language Models Fail at Tabular Prediction(arxiv.org)
12d ago·33 comments
2 pts
Knowing When to Quit: Diagnosing and Training LLMs to Abort Futile Reasoning(arxiv.org)
12d ago·discuss
1 pts
Hollow-LLM Attack: Ghost Weights That Fool Zero-Knowledge LLM Verification(arxiv.org)
12d ago·discuss
25 pts
An AI-supervised remote exam went so badly that 58,000 students must retake it(arstechnica.com)
12d ago·7 comments
1 pts
Mind the Gap: Policy vs. Reality in Post-Quantum TLS Deployment(arxiv.org)
12d ago·discuss
5 pts
AI enthusiast mods BIOS with Claude Code, defeats RSA-2048 signature checks(tomshardware.com)
13d ago·discuss
3 pts
'A waste of time for all of us': caught in the crossfire of the peer-review wars(nature.com)
13d ago·discuss
3 pts
Language Models Agree with Each Other, Not with Readers(arxiv.org)
13d ago·discuss
2 pts
Situational Awareness got the future right but misread the past(ft.com)
13d ago·discuss
2 pts
Google Just Ruined One of Its Most Important Tools(theatlantic.com)
13d ago·discuss
23 pts
Amazon completes $50B investment in OpenAI(ft.com)
13d ago·1 comments
20 pts
Coldcard wallet RNG flaw likely linked to $88M Bitcoin theft(bleepingcomputer.com)
13d ago·discuss
2 pts
Microsoft Paint used as a monitor to run Doom at up to 35 FPS(tomshardware.com)
14d ago·discuss
3 pts
First open-source firmware for AM5 launches – Coreboot and openSIL on Zen 4(tomshardware.com)
14d ago·discuss