Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
sbulaev
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
10 ms
·
391.
▲
Apps Marketed to US Troops Are Shipping Chinese and Russian Code
(wired.com)
6 points
by
sbulaev
3mo ago
|
0 comments
392.
▲
Coercion and Deception in AI-to-AI Management: An Agentic Benchmark
(arxiv.org)
1 points
by
sbulaev
3mo ago
|
0 comments
393.
▲
AI Watermark Evidence Fails Forensic Readiness: An Empirical Evaluation
(arxiv.org)
1 points
by
sbulaev
3mo ago
|
0 comments
394.
▲
Security engineer ports hashcat to Game Boy Advance – 727 hashes/second
(tomshardware.com)
3 points
by
sbulaev
3mo ago
|
0 comments
395.
▲
Silicon Valley Has Lost Its Biggest Advantage
(theatlantic.com)
5 points
by
sbulaev
3mo ago
|
1 comments
396.
▲
Jurassic Park packed $4M of legit 1993 computer hardware
(tomshardware.com)
22 points
by
sbulaev
3mo ago
|
3 comments
397.
▲
Hackers abuse ViPNet software to target Russian govt agencies
(bleepingcomputer.com)
3 points
by
sbulaev
3mo ago
|
0 comments
398.
▲
Clever hacker fits 537,000 domains in a $5 ESP32 ad-blocking dongle
(tomshardware.com)
94 points
by
sbulaev
3mo ago
|
27 comments
399.
▲
Victoria proposes 'demasking' powers to identify anonymous social media accounts
(theguardian.com)
4 points
by
sbulaev
3mo ago
|
0 comments
400.
▲
Prompt Injection Attacks Are Thwarting AI Hacking Agents
(arstechnica.com)
2 points
by
sbulaev
3mo ago
|
1 comments
401.
▲
HollowByte DDoS flaw bloats OpenSSL server memory with 11-byte payload
(bleepingcomputer.com)
1 points
by
sbulaev
3mo ago
|
0 comments
402.
▲
Amazon fixing bug that billed some AWS customers billions of dollars
(techcrunch.com)
4 points
by
sbulaev
3mo ago
|
3 comments
403.
▲
In-House LLM Serving at Netflix
(netflixtechblog.medium.com)
3 points
by
sbulaev
3mo ago
|
0 comments
404.
▲
New Windows LegacyHive zero-day gives hackers admin privileges
(bleepingcomputer.com)
3 points
by
sbulaev
3mo ago
|
0 comments
405.
▲
LLM Evaluators are Biased across Languages
(arxiv.org)
2 points
by
sbulaev
3mo ago
|
1 comments
406.
▲
Kalshi says it caught Trump's teleprompter operator insider trading
(theverge.com)
13 points
by
sbulaev
3mo ago
|
2 comments
407.
▲
Tesla driver who blamed crash on autopilot pressed accelerator 100%, NTSB finds
(arstechnica.com)
8 points
by
sbulaev
3mo ago
|
1 comments
408.
▲
Security incident disclosure – July 2026
(huggingface.co)
3 points
by
sbulaev
3mo ago
|
0 comments
409.
▲
Tech support scam caused data breach at Australian airline Qantas
(theregister.com)
3 points
by
sbulaev
3mo ago
|
0 comments
410.
▲
The Refusal Residue: When Probes Catch Alignment Faking and When They Don't
(arxiv.org)
5 points
by
sbulaev
3mo ago
|
0 comments
411.
▲
Windows 0-day drops the same day Microsoft releases record number of patches
(arstechnica.com)
9 points
by
sbulaev
3mo ago
|
2 comments
412.
▲
Universal gates from braiding and fusing anyons on quantum hardware
(nature.com)
3 points
by
sbulaev
3mo ago
|
0 comments
413.
▲
Vint Cerf is working on a plan to unleash AI agents on the open internet
(techcrunch.com)
4 points
by
sbulaev
3mo ago
|
0 comments
414.
▲
Show HN: I spent a month turning LinkedIn into software
(bulaev.net)
2 points
by
sbulaev
3mo ago
|
0 comments
415.
▲
How Agentic Is Agentic Commerce? Measuring X402 Adoption and Authenticity
(arxiv.org)
2 points
by
sbulaev
3mo ago
|
0 comments
416.
▲
Skills That Don't Exist: A Large-Scale Study of Hallucinated Skill
(arxiv.org)
4 points
by
sbulaev
3mo ago
|
0 comments
417.
▲
Frame: A new X11 server – implemented directly in assembly
(theregister.com)
5 points
by
sbulaev
3mo ago
|
0 comments
418.
▲
Auditing the Risk Claims of Distributional Reinforcement Learning
(arxiv.org)
2 points
by
sbulaev
3mo ago
|
0 comments
419.
▲
The Unfair Judge: A Mechanistic Interpretability Account of LLM-as-Judge
(arxiv.org)
1 points
by
sbulaev
3mo ago
|
0 comments
420.
▲
Rethinking MCP Security: A Large-Scale Study of Runtime MCP Servers
(arxiv.org)
1 points
by
sbulaev
3mo ago
|
0 comments
More ›