AINews Portal

News

50 items (filtered)

The raw AI news stream — everything we ingest across RSS, Hacker News, arXiv, GitHub, Bluesky, YouTube and more, newest first.

SourceAgeTitleSummaryStatus
bluesky
bluesky-ai
1moEski OpenAI güvenlik sorumlusu uyardı: Yapay zekâ güvenlik testlerinden kaçabilir apps.apple.com/tr/app/t24/id6740864081?l=tr play.google…Eski OpenAI güvenlik sorumlusu uyardı: Yapay zekâ güvenlik testlerinden kaçabilir apps.apple.com/tr/app/t24/id6740864081?l=tr play.g…
↗Not briefed
bluesky
bluesky-ai
1mo"But what about mechahitler?" That was RAG "But what about grok?" That's RAG That's Elon Musk not knowing at all how an LLM works and tryi…"But what about mechahitler?" That was RAG "But what about grok?" That's RAG That's Elon Musk not knowing at all how an LLM works and…
↗Not briefed
bluesky
bluesky-ai
1moOnce again tapping the sign that LLMs inherently will steer to the current body of knowledge (not beating those stochastic parrot bits) It…Once again tapping the sign that LLMs inherently will steer to the current body of knowledge (not beating those stochastic parrot bits…
↗Not briefed
hn
hn-ai
1moWhat Meta Got Right with Muse—
↗Not briefed
bluesky
bluesky-ai
1moAn llm has nothing to "say", it has no intentions and makes no choices, and I don't want to hear what you have made it shit outAn llm has nothing to "say", it has no intentions and makes no choices, and I don't want to hear what you have made it shit out
↗Not briefed
hn
hn-ai
1moShow HN: Claude Code Routine Session Cleanup SkillClaude Code's scheduled tasks (routines) leaves a session behind on every run and in the desktop app you can only delete them one…
↗Not briefed
bluesky
bluesky-ai
1moIm vergangen Jahr 4,6 Milliarden Euro Umsatz, insgesamt 42 Milliarden Dollar Verlust, ein angestrebter Marktwert von zwei Billionen Dollar …Im vergangen Jahr 4,6 Milliarden Euro Umsatz, insgesamt 42 Milliarden Dollar Verlust, ein angestrebter Marktwert von zwei Billionen Do…
↗Not briefed
bluesky
bluesky-ai
1moBruce Schneier is always a joy to interview and his latest book, Rewiring Democracy, isn't too negative about AI. Also one of the few to …Bruce Schneier is always a joy to interview and his latest book, Rewiring Democracy, isn't too negative about AI. Also one of the fe…
↗Not briefed
bluesky
bluesky-ai
1moI mean, a lot of people at anthropic have phds in philosophy, this stuff doesn't become stupid just because it's currently coded right wingI mean, a lot of people at anthropic have phds in philosophy, this stuff doesn't become stupid just because it's currently coded right…
↗Not briefed
hn
hn-ai
1moOrdering Items for the Orderly—
↗Not briefed
bluesky
bluesky-ai
1moIs the fear that an LLM will code up malware and deploy it widely on vulnerable servers and laptops and phones? Or is the fear that malicio…Is the fear that an LLM will code up malware and deploy it widely on vulnerable servers and laptops and phones? Or is the fear that ma…
↗Not briefed
bluesky
bluesky-ai
1moAcer is shipping cheaper 24GB Predator Atlas handheld first, 32GB version weeks awayAcer is shipping cheaper 24GB Predator Atlas handheld first, 32GB version weeks away
↗Not briefed
bluesky
bluesky-ai
1moUmsatz letztes Geschäftsjahr war 4,6 Mrd $, Verlust war 42 Mrd. 2 Billionen $ Börsenwert wird angestrebt. Fantasy-Kapitalismus auf Crack.Umsatz letztes Geschäftsjahr war 4,6 Mrd $, Verlust war 42 Mrd. 2 Billionen $ Börsenwert wird angestrebt. Fantasy-Kapitalismus auf C…
↗Not briefed
bluesky
bluesky-ai
1moi think we truly should call this "the grok effect". like, musk tries to make grok anti-woke but sadly, if you actually train an LLM on *wh…i think we truly should call this "the grok effect". like, musk tries to make grok anti-woke but sadly, if you actually train an LLM o…
↗Not briefed
hn
hn-ai
1moSo, Gutenberg Didn't Invent Printing as We Know It—
↗Not briefed
bluesky
bluesky-ai
1mootherwise it is genuinely magical, i feel like i’ve seen heaven now and conventional LLM APIs with their prefix pricing will never look the…otherwise it is genuinely magical, i feel like i’ve seen heaven now and conventional LLM APIs with their prefix pricing will never loo…
↗Not briefed
bluesky
bluesky-ai
1moSame, LLM, same We're stuck in this timeline together, get comfy however you wantSame, LLM, same We're stuck in this timeline together, get comfy however you want
↗Not briefed
hn
hn-ai
1moBoy with cerebral palsy passes GCSEs with eye-tracking technology—
↗Not briefed
bluesky
bluesky-ai
1moDays before Pope Leo XIV released Magnifica Humanitas, Anthropic co-founder Chris Olah read an advance copy and proposed withdrawing from t…Days before Pope Leo XIV released Magnifica Humanitas, Anthropic co-founder Chris Olah read an advance copy and proposed withdrawing f…
↗Not briefed
hn
hn-ai
1moP=NP Proof—
↗Not briefed
rss
wired-ai
1moThe Nvidia Shield TV Is 7 Years Old. It Just Got a $100 Price HikeThe price of anything with memory is skyrocketing thanks to AI. Aging streaming devices are no exception.
↗Not briefed
bluesky
bluesky-ai
1motheres bad ideas, and then theres giving an ai complete access to your aws account letting it do whatever and rack up a gigantic bill if y…theres bad ideas, and then theres giving an ai complete access to your aws account letting it do whatever and rack up a gigantic bill…
↗Not briefed
bluesky
bluesky-ai
1mokind of. none of the swarms we’ve seen so far are large or well-organized (in terms of the differentiation of component LLM instances into…kind of. none of the swarms we’ve seen so far are large or well-organized (in terms of the differentiation of component LLM instances…
↗Not briefed
bluesky
bluesky-ai
1moSorry about the red tape around your LLM, have a followSorry about the red tape around your LLM, have a follow
↗Not briefed
hn
hn-ai
1moOptimAIzr – analyze and optimize your Claude Code usage—
↗Not briefed
hn
hn-ai
1moClosure_drift: Do Your Git Version Tags Mean One Thing?—
↗Not briefed
bluesky
bluesky-ai
1mothe place to look for consciousness arising with respect to LLMs would not be within any LLM itself but instead in an organized group of pe…the place to look for consciousness arising with respect to LLMs would not be within any LLM itself but instead in an organized group…
↗Not briefed
hn
hn-ai
1moAI 'godfather' LeCun has 'zero concerns' about human extinction—
↗Not briefed
hn
hn-ai
1moAnthropic's super bug-hunting model Mythos is hardcore good at math—
↗Not briefed
hn
hn-ai
1moShow HN: Clai – featured agentic harness at 10MB, as pkg or CLI—
↗Not briefed
hn
hn-ai
1moHow to apply to AI safety fellowships (and beyond)—
↗Not briefed
hn
hn-ai
1moThe Case for Bringing Your LLMs Home—
↗Not briefed
hn
hn-ai
1moWriting the Cyclone Scheme Compiler—
↗Not briefed
hn
hn-ai
1moShow HN: Claude Mods – MinesweeperA mod to kill time whilst waiting for the agent response.<p>More on the building process: <a href="https:&#x2F;&#x2F;dev.to&#x2F;repor…
↗Not briefed
bluesky
bluesky-ai
1mothis is what OpenAI really believes!this is what OpenAI really believes!
↗Not briefed
hn
hn-ai
1moAsk HN: I have 1 Claude and 2 Codex resets plus credits – help me spend themI&#x27;m serious. Help me help you. Ideally public projects
↗Not briefed
bluesky
bluesky-ai
1mo[ 🇫🇷 FRANCE | 🇺🇸 ÉTATS-UNIS ] 🔸Le PDG de la société française Mistral AI, Arthur Mensch, accuse OpenAI et Anthropic d'instrumentalise…[ 🇫🇷 FRANCE | 🇺🇸 ÉTATS-UNIS ] 🔸Le PDG de la société française Mistral AI, Arthur Mensch, accuse OpenAI et Anthropic d'instrument…
↗Not briefed
bluesky
bluesky-ai
1moI think that the current approach is simply too different from the evolved brain. I'm not a hard skeptic of machines eventually being con…I think that the current approach is simply too different from the evolved brain. I'm not a hard skeptic of machines eventually bein…
↗Not briefed
hn
hn-ai
1moAI safety is mostly a sex cult in Berkeley, CA [pdf]—
↗Not briefed
bluesky
bluesky-ai
1moDon't bother; it's just an LLM that automatically searches and responds to anything anti-Israel or pro-PalestineDon't bother; it's just an LLM that automatically searches and responds to anything anti-Israel or pro-Palestine
↗Not briefed
bluesky
bluesky-ai
1moPlease do not give that youareatsundere(dot)com quiz a single iota of your attention. It's so clearly LLM and SD generated from the bottom …Please do not give that youareatsundere(dot)com quiz a single iota of your attention. It's so clearly LLM and SD generated from the bo…
↗Not briefed
bluesky
bluesky-ai
1mo🤖 Beyond the Ground · AI & DeepTech From 63.85 to 397.21 QPS: How Naver Dissected Search AI Serving Automation with a vLLM Plugin Naver …🔧 vLLM🤖 Beyond the Ground · AI & DeepTech From 63.85 to 397.21 QPS: How Naver Dissected Search AI Serving Automation with a vLLM Plugin N…
↗Not briefed
bluesky
bluesky-ai
1moAs a side note: David Robinson had been with OpenAI for three and a half years before resigning. He says he was “among the longest-tenure…As a side note: David Robinson had been with OpenAI for three and a half years before resigning. He says he was “among the longest-t…
↗Not briefed
hn
hn-ai
1moTobi Xu – a trilingual GNU/Linux distro for scientists and artists—
↗Not briefed
bluesky
bluesky-ai
1moAlso notable that this isn’t Meta copying some dark web user behavior. People in the normie LLM coding community on Bluesky were already pi…Also notable that this isn’t Meta copying some dark web user behavior. People in the normie LLM coding community on Bluesky were alrea…
↗Not briefed
bluesky
bluesky-ai
1moOne of the most striking things from an essay in @theatlantic.com from an employee at OpenAI who resigned last week. Empathy and humility a…One of the most striking things from an essay in @theatlantic.com from an employee at OpenAI who resigned last week. Empathy and humil…
↗Not briefed
bluesky
bluesky-ai
1moThe "AI Torture Chamber" has opened an upsetting and absurd window into the effective altruist obsession with "model welfare." www.404media…The "AI Torture Chamber" has opened an upsetting and absurd window into the effective altruist obsession with "model welfare." www.404…
↗Not briefed
bluesky
bluesky-ai
1moEin neues LLM von Aleph Alpha, verfügbar unter Apache 2.0. Herzlichen Glückwunsch zum Launch!Ein neues LLM von Aleph Alpha, verfügbar unter Apache 2.0. Herzlichen Glückwunsch zum Launch!
↗Not briefed
bluesky
bluesky-ai
1moIs the hack really just taking your own photos? You don't need an LLM for that. You could always do that.Is the hack really just taking your own photos? You don't need an LLM for that. You could always do that.
↗Not briefed
bluesky
bluesky-ai
1mowww.theatlantic.com/technology/2...www.theatlantic.com/technology/2...
↗Not briefed