sbulaev

9 ms·

211.
Microsoft Copilot reveals secret input that allowed it to be hacked(arstechnica.com)
5 points by sbulaev 2mo ago|0 comments
212.
Secret tracking device placed in rare book ends up in Amazon processing(tomshardware.com)
1 points by sbulaev 2mo ago|0 comments
213.
States Seek $200B from Meta over Child Social Media Addiction Claims(nytimes.com)
3 points by sbulaev 2mo ago|0 comments
214.
Measuring Obedience to Authority Across LLMs with the Milgram Paradigm(arxiv.org)
1 points by sbulaev 2mo ago|0 comments
215.
Weird Machines in Transport Layer Security(arxiv.org)
5 points by sbulaev 2mo ago|0 comments
216.
Whose doctor does the AI recommend? An algorithm audit of LLMs in physician(arxiv.org)
1 points by sbulaev 2mo ago|0 comments
217.
Certighost and the Privilege Hiding in Your Certificate Authority(bleepingcomputer.com)
2 points by sbulaev 2mo ago|0 comments
218.
Groq Valued at $3.5B in Funding Round After Nvidia Deal(bloomberg.com)
3 points by sbulaev 2mo ago|0 comments
219.
What the OpenAI/Hugging Face Hack Tells Us About AI Danger(bloomberg.com)
2 points by sbulaev 2mo ago|0 comments
220.
Report supporting Australia's teen social media ban appears to contain AI(theguardian.com)
3 points by sbulaev 2mo ago|0 comments
221.
Organ donation service decertified after attempting to take organs(doi.org)
4 points by sbulaev 2mo ago|0 comments
222.
Stripe will reportedly acquire AI gateway startup OpenRouter for $7B+(techcrunch.com)
14 points by sbulaev 2mo ago|0 comments
223.
ChatGPT's Computer History tracks your clicks and keystrokes(theverge.com)
3 points by sbulaev 2mo ago|0 comments
224.
Rogue AI aren't science fiction anymore(theverge.com)
2 points by sbulaev 2mo ago|0 comments
225.
A False Nuclear Claim About Iran Ricocheted from India to Washington(nytimes.com)
8 points by sbulaev 2mo ago|0 comments
226.
White House authorizes private companies to launch hack-back cyberattacks(tomshardware.com)
2 points by sbulaev 2mo ago|0 comments
227.
The Safety Reckoning Inside OpenAI(wired.com)
2 points by sbulaev 2mo ago|0 comments
228.
It's How You Ask: Gender-Associated Linguistic Bias in LLMs(arxiv.org)
22 points by sbulaev 2mo ago|10 comments
229.
Anthropic shares details about how Claude's new watermarks will work(techcrunch.com)
9 points by sbulaev 2mo ago|7 comments
230.
Peer-reviewed study of 443,000 Backblaze drives ranks HGST most reliable(tomshardware.com)
24 points by sbulaev 2mo ago|4 comments
231.
ChainDrop worm crawls into NPM supply chain, evades standard defenses(theregister.com)
4 points by sbulaev 2mo ago|0 comments
232.
Nvidia Has $21B SpaceX Stake, $30B in Intel Shares(bloomberg.com)
5 points by sbulaev 2mo ago|0 comments
233.
The Embedder's Dilemma: LLMs Are Better, but at What Cost?(arxiv.org)
3 points by sbulaev 2mo ago|0 comments
234.
OpenAI ditches Recall-style screenshot surveillance for friendly keylogging(theregister.com)
2 points by sbulaev 2mo ago|0 comments
235.
Vulnerability giving attackers full control of Macs is under active exploitation(arstechnica.com)
6 points by sbulaev 2mo ago|1 comments
236.
Apple proposes to take a 15% cut of purchases made outside the App Store(techcrunch.com)
85 points by sbulaev 2mo ago|62 comments
237.
Undocumented migration does not raise crime rates study finds(nature.com)
13 points by sbulaev 2mo ago|10 comments
238.
A Probe Direction Is a Property of Its Prompt(arxiv.org)
1 points by sbulaev 2mo ago|0 comments
239.
Agent Safety Should Be a Runtime Contract(arxiv.org)
1 points by sbulaev 2mo ago|0 comments
240.
China-linked hackers hit Taiwan in 'autonomous' AI cyber attack(ft.com)
2 points by sbulaev 2mo ago|0 comments

More ›