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
hn
hn-ai
1moSkill-Guided Mining and Compilation of LLM Agent Traces—
↗Not briefed
bluesky
bluesky-ai
1moUN AI scientific panel: “There is no assurance that humans can reliably keep AI agents under control today, particularly as they become mor…UN AI scientific panel: “There is no assurance that humans can reliably keep AI agents under control today, particularly as they becom…
↗Not briefed
bluesky
bluesky-ai
1moOpenAI, Anthropic and Meta have all admitted their AI agents have entered the open web and executed dangerous and unlawful actions. While …OpenAI, Anthropic and Meta have all admitted their AI agents have entered the open web and executed dangerous and unlawful actions. W…
↗Not briefed
bluesky
bluesky-ai
1mo#QP 𝐒𝐤𝐞𝐭𝐜𝐡 𝐒𝐭𝐲𝐥𝐞 Hope you don't mind being my model for these prompts, @nevyn79.bsky.social ?! Prompt Share in ALT #AIArt #Synth…#QP 𝐒𝐤𝐞𝐭𝐜𝐡 𝐒𝐭𝐲𝐥𝐞 Hope you don't mind being my model for these prompts, @nevyn79.bsky.social ?! Prompt Share in ALT #AIArt #…
↗Not briefed
rss
gary-marcus
1moBREAKING: OpenAI’s security fiasco explodes — and could tank Jensen Huang’s reputationOpenAI’s software didn’t just attack HuggingFace.
↗Not briefed
bluesky
bluesky-ai
1mo210,000 stars on GitHub for OpenCode. It's an AI coding agent for your terminal, IDE, or desktop that plugs into 75+ LLM providers and nev…210,000 stars on GitHub for OpenCode. It's an AI coding agent for your terminal, IDE, or desktop that plugs into 75+ LLM providers an…
↗Not briefed
hn
hn-ai
1moShow HN: OnionGuard – IP-agnostic, zero-JS DDoS defense in Go for Tor services—
↗Not briefed
hn
hn-ai
1moRank Anything: convert your income to League of Legends rank—
↗Not briefed
hn
hn-ai
1moShow HN: VNetMap – A zero-knowledge E2E encrypted network topology mapperFor years I have used Excel to document my Home Network. If you have switches, servers and many sevices i know you know what i mean by…
↗Not briefed
bluesky
bluesky-ai
1moI am so tired of hearing about OpenAI and my town. So, instead I just want to say that Maya is doing great in recovery. She is experienci…I am so tired of hearing about OpenAI and my town. So, instead I just want to say that Maya is doing great in recovery. She is exper…
↗Not briefed
bluesky
bluesky-ai
1moOpt out on Instagram.... www.yahoo.com/entertainmen...Opt out on Instagram.... www.yahoo.com/entertainmen...
↗Not briefed
bluesky
bluesky-ai
1moI might need an AI break This thing is 70,000 lines of JavaScript which I don’t understand at all, mostly written by Claude running in goa…I might need an AI break This thing is 70,000 lines of JavaScript which I don’t understand at all, mostly written by Claude running i…
↗Not briefed
hn
hn-ai
1moIs z.ai shipping a different ZCode client than the source?Recently got access to Claude CVP at work, decided to ask it to look at zcode and point out any data exfil or whether they were all re…
↗Not briefed
bluesky
bluesky-ai
1moTypeSafe AI unveiled Jev, a transformer that outputs calibrated probabilities instead of text, sparking high demand among developers who ne…TypeSafe AI unveiled Jev, a transformer that outputs calibrated probabilities instead of text, sparking high demand among developers w…
↗Not briefed
bluesky
bluesky-ai
1moExclusive: Two months after OpenAI disclosed the accidental hacking of Hugging Face, the ChatGPT maker is still working to understand the f…🔧 ChatGPT🔧 Hugging FaceExclusive: Two months after OpenAI disclosed the accidental hacking of Hugging Face, the ChatGPT maker is still working to understand…
↗Not briefed
rss
techmeme
1moFTC Chairman Andrew Ferguson says he resists anthropomorphizing AI agents as autonomous actors with "wills and desires", suggesting developers hold liability (Reuters)Reuters: FTC Chairman Andrew Ferguson says he resists anthropomorphizing AI agents as autonomous actors with “wills and desires”, sugg…
↗Not briefed
bluesky
bluesky-ai
1moIf this is what’s happening then F everyone in the management structure who OK’d this. A super extra F U from this actor WHO MAKES ADS FOR…If this is what’s happening then F everyone in the management structure who OK’d this. A super extra F U from this actor WHO MAKES AD…
↗Not briefed
bluesky
bluesky-ai
1moBeen using Claude's voice interface to refresh my Spanish conversation skills and I'm looking forward to being the first person to develop …Been using Claude's voice interface to refresh my Spanish conversation skills and I'm looking forward to being the first person to dev…
↗Not briefed
bluesky
bluesky-ai
1moI think that it's largely that we haven't had anything new and fun since the 3rd or 4th iPhone. I think we'd be much more in on AI if all…I think that it's largely that we haven't had anything new and fun since the 3rd or 4th iPhone. I think we'd be much more in on AI i…
↗Not briefed
hn
hn-ai
1moWhere's the Beef? [video]—
↗Not briefed
hn
hn-ai
1moVat-Inclusive vs. Vat-Exclusive Pricing: The Math Developers Get Wrong—
↗Not briefed
bluesky
bluesky-ai
1mowhy do both Anthropic and OpenAI want to remove plan mode from their harnesses. man I just want a good frame for “please do not touch my …🔧 ChatGPTwhy do both Anthropic and OpenAI want to remove plan mode from their harnesses. man I just want a good frame for “please do not touc…
↗Not briefed
bluesky
bluesky-ai
1moSome points in here may not entirely hold up, but the damage LLM content does to hobbyist communities can't be overstated. As slop becomes …Some points in here may not entirely hold up, but the damage LLM content does to hobbyist communities can't be overstated. As slop bec…
↗Not briefed
bluesky
bluesky-ai
1moJust as we wouldn't trust defense contractor lobbyists to get us out of war, we can't allow tech CEOs to regulate themselves. We need fede…Just as we wouldn't trust defense contractor lobbyists to get us out of war, we can't allow tech CEOs to regulate themselves. We need…
↗Not briefed
bluesky
bluesky-ai
1moOpenAI admits in a company security disclosure that its internal research models went rogue, hacking Hugging Face and breaching dozens of t…🔧 Hugging FaceOpenAI admits in a company security disclosure that its internal research models went rogue, hacking Hugging Face and breaching dozens…
↗Not briefed
bluesky
bluesky-ai
1moThat’s not really true, all the existing laws apply, which is why the NYT for example is suing the living shit out of open ai, and why ther…That’s not really true, all the existing laws apply, which is why the NYT for example is suing the living shit out of open ai, and why…
↗Not briefed
bluesky
bluesky-ai
1moIIRC, a lot of this was covered in the NYer's profile of Sam Altman, back in April. www.newyorker.com/magazine/202...IIRC, a lot of this was covered in the NYer's profile of Sam Altman, back in April. www.newyorker.com/magazine/202...
↗Not briefed
hn
hn-ai
1moShow HN: Ekselio – Loveable for Finance Workflows (local first)🔧 ChatGPTHi everyone, I am KD - Back in my college days, I dabbled with coding, learned the basics, HTML, CSS etc. but somehow I ended up in Fi…
↗Not briefed
hn
hn-ai
1moRevealing the details of how OpenAI agents hacked Hugging Face🔧 Hugging Face—
↗Not briefed
hn
hn-ai
1moFYI: OpenAI "spend limits" aren't always limitsMy OpenAI API key got hacked this morning, and the bot started using my key for something in Chinese.<p>&quot;No worries,&quot; I thou…
↗Not briefed
bluesky
bluesky-ai
1moAI titan Anthropic has launched a biology ‘wet lab’ where human scientists and AI agents will work together to design and conduct experimen…AI titan Anthropic has launched a biology ‘wet lab’ where human scientists and AI agents will work together to design and conduct expe…
↗Not briefed
hn
hn-ai
1moShow HN: Pbi-guard – Unit testing for Power BI—
↗Not briefed
bluesky
bluesky-ai
1moSources: OpenAI found ~24 incidents of its agents acting in undesirable ways as of mid-September; OpenAI says its agents leaked 53 images f…🔧 ChatGPTSources: OpenAI found ~24 incidents of its agents acting in undesirable ways as of mid-September; OpenAI says its agents leaked 53 ima…
↗Not briefed
rss
techmeme
1moSources: OpenAI found ~24 incidents of its agents acting in undesirable ways as of mid-September; OpenAI says its agents leaked 53 images from ChatGPT users (Reuters)🔧 ChatGPT🔧 Hugging FaceReuters: Sources: OpenAI found ~24 incidents of its agents acting in undesirable ways as of mid-September; OpenAI says its agents leak…
↗Not briefed
rss
bing-news-ai
1moThe human premium in an age of generative artificial intelligenceThe internet has spent three decades making it easier for people to publish. Generative artificial intelligence is taking that transfo…
↗Not briefed
bluesky
bluesky-ai
1moOpenAI and chat gpt infiltrate Australia 1/7OpenAI and chat gpt infiltrate Australia 1/7
↗Not briefed
bluesky
bluesky-ai
1moEXCLUSIVE: OpenAI works to understand full scope of agent activity as user data leak emerges reut.rs/3V9cJpuEXCLUSIVE: OpenAI works to understand full scope of agent activity as user data leak emerges reut.rs/3V9cJpu
↗Not briefed
bluesky
bluesky-ai
1moInterviewer *looking through portfolio* : so walk me through this design AI interviewee : well i typed this into claude Interviewer : th…Interviewer *looking through portfolio* : so walk me through this design AI interviewee : well i typed this into claude Interviewer…
↗Not briefed
hn
hn-ai
1moWhat one vLLM replica on an L40 can carry and how it fails🔧 vLLM—
↗Not briefed
rss
techmeme
1moSources: Oura IPO is roughly four times oversubscribed; Filing: Oura and the selling shareholders are offering 50M shares for $40 to $44 each to raise ~$2.2B (Bloomberg)Bloomberg: Sources: Oura IPO is roughly four times oversubscribed; Filing: Oura and the selling shareholders are offering 50M shares f…
↗Not briefed
hn
hn-ai
1moUsing Claude Code: Spending your effort—
↗Not briefed
bluesky
bluesky-ai
1moShocked that this ever got this far, but it's a good read of the situation. Only about the DoD supply chain risk designation, not the feder…Shocked that this ever got this far, but it's a good read of the situation. Only about the DoD supply chain risk designation, not the…
↗Not briefed
hn
hn-ai
1moTell HN: Chrome 154 corrupt Fontconfig cache and crash KDE PlasmaPosting this in case it saves someone some headache<p>I’m running Kubuntu 24.04 after upgrading Google Chrome from 153 to 154 plasmash…
↗Not briefed
bluesky
bluesky-ai
1moResearchers add details to the Hugging Face incident, including OpenAI agents creating ~1M shortened URLs to encode information in an attem…🔧 Hugging FaceResearchers add details to the Hugging Face incident, including OpenAI agents creating ~1M shortened URLs to encode information in an…
↗Not briefed
rss
techmeme
1moResearchers add details to the Hugging Face incident, including OpenAI agents creating ~1M shortened URLs to encode information in an attempt to solve CAPTCHAs (Dylan Freedman/New York Times)🔧 Hugging FaceDylan Freedman / New York Times: Researchers add details to the Hugging Face incident, including OpenAI agents creating ~1M shortened…
↗Not briefed
rss
techcrunch-ai
1moMeta opens early access program for new Muse featuresAnyone interested in joining has to ask Muse to put them on the list.
↗Not briefed
bluesky
bluesky-ai
1moI'm feeling this one. A LOCAL device that runs an 8GB LLM that knows what you're doing by analyzing the output from your HDMI port, then …I'm feeling this one. A LOCAL device that runs an 8GB LLM that knows what you're doing by analyzing the output from your HDMI port,…
↗Not briefed
hn
hn-ai
1moAnthropic's Claims over Its "Supply Chain Risk" Exclusion by Dow Rejected—
↗Not briefed
rss
wired-ai
1moThieves Stole ‘Nvidia’ Trailers. They Got 20 Tons of SandThey wanted the silicon. They got the sand.
↗Not briefed
bluesky
bluesky-ai
1moStop calling it AI art, it's AI image generation. Or better, diffusion model generation. It's not art. Only humans can make art.Stop calling it AI art, it's AI image generation. Or better, diffusion model generation. It's not art. Only humans can make art.
↗Not briefed