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
19hAsk HN: How do you protect self-hosted websites from LLM scrapers?Hello HN,<p>I&#x27;m considering self-hosting a couple of static low-volume traffic on a VM. I&#x27;m curious to know what techniques…
Not briefed
rss
techmeme
19hGoogle starts rolling out the google.com/goto URL as a passthrough URL to help prevent scraping of its search results by third-party tools and AI companies (Barry Schwartz/Search Engine Roundtable)Barry Schwartz / Search Engine Roundtable: Google starts rolling out the google.com/goto URL as a passthrough URL to help prevent scra…
Not briefed
hn
hn-ai
19hShow HN: Wallps: set different live wallpapers for macOS desktop and lock screenmacOS ties the lock screen to your desktop wallpaper, with no built-in option to set them separately. Wallps decouples the two, lettin…
Not briefed
bluesky
bluesky-ai
19hTHEY DID THIS WITH A FUCKING FLASH MODEL!? Oh yeah it is so game over for the current pricings of Anthropic and OpenAI's models, unless th…THEY DID THIS WITH A FUCKING FLASH MODEL!? Oh yeah it is so game over for the current pricings of Anthropic and OpenAI's models, unle…
Not briefed
rss
techmeme
19hArturo Béjar, a key witness in state AGs' trial against Meta, says terms of the settlement are insufficient to protect young users; Meta rejects the criticism (Nick Robins-Early/The Guardian)Nick Robins-Early / The Guardian: Arturo Béjar, a key witness in state AGs' trial against Meta, says terms of the settlement are insuf…
Not briefed
bluesky
bluesky-ai
19hhow much of this do you think is them fearing more Jalapeño-esque developments? like in theory if OpenAI or Anthropic start emailing thei…how much of this do you think is them fearing more Jalapeño-esque developments? like in theory if OpenAI or Anthropic start emailing…
Not briefed
bluesky
bluesky-ai
19hIt does kinda read like hugging face wasn't the only thing openAI compromised and they're in NDAs with the other targets, right?🔧 Hugging FaceIt does kinda read like hugging face wasn't the only thing openAI compromised and they're in NDAs with the other targets, right?
Not briefed
rss
techmeme
19hFiling: Nvidia says it has $18B committed to equity investments for the rest of the fiscal year and held $47.9B in private companies as of late July (Geoff Weiss/Business Insider)Geoff Weiss / Business Insider: Filing: Nvidia says it has $18B committed to equity investments for the rest of the fiscal year and he…
Not briefed
bluesky
bluesky-ai
19hOpenAI has reported that russian operators used ChatGPT to conduct a covert influence campaign aimed at promoting pro-russian narratives an…🔧 ChatGPTOpenAI has reported that russian operators used ChatGPT to conduct a covert influence campaign aimed at promoting pro-russian narrativ…
Not briefed
rss
techmeme
19hMeta seeks to position itself as the industry leader in child safety through its settlement with US AGs, while still refusing to admit any wrongdoing (Casey Newton/Platformer)Casey Newton / Platformer: Meta seeks to position itself as the industry leader in child safety through its settlement with US AGs, wh…
Not briefed
bluesky
bluesky-ai
19hI don’t think anyone but me cares but this is not about an LLMI don’t think anyone but me cares but this is not about an LLM
Not briefed
rss
techmeme
19hSources: the US is probing Singapore-based Apex Logistics for its suspected role in smuggling servers with Nvidia AI chips to China; Apex says it is cooperating (Bloomberg)Bloomberg: Sources: the US is probing Singapore-based Apex Logistics for its suspected role in smuggling servers with Nvidia AI chips…
Not briefed
hn
hn-ai
19hShow HN: YASS – Open-source stock screener with point-in-time backtesting
Not briefed
rss
techmeme
19hMeta's AG settlement gives it a moat to write child safety rules for other social media platforms, with little evidence its new safety features will help kids (Mike Masnick/Techdirt)Mike Masnick / Techdirt: Meta's AG settlement gives it a moat to write child safety rules for other social media platforms, with littl…
Not briefed
bluesky
bluesky-ai
19hThe primary driver was that OpenAI were arrogantThe primary driver was that OpenAI were arrogant
Not briefed
hn
hn-ai
19hAsk HN: Would you pay for an AI API that guarantees zero prompt retention?i am just tired of looking for a reliable open-source model AI API with ZDR and don&#x27;t even save my prompts. I&#x27;m thinking of…
Not briefed
bluesky
bluesky-ai
19hwww.commondreams.org/news/openai-... “It’s a bad idea to have our society run by people like this who are so far out of touch with reality …www.commondreams.org/news/openai-... “It’s a bad idea to have our society run by people like this who are so far out of touch with rea…
Not briefed
hn
hn-ai
19hShow HN: Historical daily weather for any US ZIP code Python Package
Not briefed
bluesky
bluesky-ai
20hsome people are just incredibly unfunny. for a second I thought this was the text that the archive generates automatically but someone actu…some people are just incredibly unfunny. for a second I thought this was the text that the archive generates automatically but someone…
Not briefed
bluesky
bluesky-ai
20hIt is correct that this will crush OpenAI and Anthropic but I also don't think it will end well for Nvidia. They don't own any actual semic…It is correct that this will crush OpenAI and Anthropic but I also don't think it will end well for Nvidia. They don't own any actual…
Not briefed
hn
hn-ai
20hShow HN: MLSentinel – monitor ML models and catch failures earlyI built MLSentinel to make it easier to monitor ML models after they are deployed.<p>One problem I kept running into was that a model…
Not briefed
bluesky
bluesky-ai
20hanti-union llmanti-union llm
Not briefed
bluesky
bluesky-ai
20hstanding outside the anthropic offices with a sign that says "failure to plan is planning to fail"standing outside the anthropic offices with a sign that says "failure to plan is planning to fail"
Not briefed
bluesky
bluesky-ai
20hHere is a fun prompt injection vector you can have for free. They've tried to train the current SOTA LLMs to detect and ignore prompt inje…Here is a fun prompt injection vector you can have for free. They've tried to train the current SOTA LLMs to detect and ignore prompt…
Not briefed
bluesky
bluesky-ai
20hOne user posting from Beijing said the model is amazing, delivering strong performance at a low cost. Another from Chongqing said that Ox A…One user posting from Beijing said the model is amazing, delivering strong performance at a low cost. Another from Chongqing said that…
Not briefed
hn
hn-ai
20hDon't Let LLMs Play Telephone with Your Ideas
Not briefed
hn
hn-ai
20hWhy LLMs Can't Play Chess
Not briefed
bluesky
bluesky-ai
20hAltman's a liar but the proof will be in the pudding and if the model works, then it works. Anthropic is going to have to start opening up…Altman's a liar but the proof will be in the pudding and if the model works, then it works. Anthropic is going to have to start openi…
Not briefed
hn
hn-ai
20hShow HN: I Designed a 3D-Printed Laminar Flow Nozzle
Not briefed
rss
marktechpost
20hGoogle Research Introduces GlucoFM: A 0.72M-Parameter Dual-Stream Foundation Model for Continuous Glucose MonitoringGoogle Research and UNSW Sydney released GlucoFM, a self-supervised foundation model that splits a CGM trace into a slow physiological…
Not briefed
hn
hn-ai
20hLlwdc – Direct-to-fan storefront for tracks, stems, and sound packs
Not briefed
bluesky
bluesky-ai
20hI think of the couple thousand beloved NES, Genesis, and SNES games that were translated messily, for a pittance, and then look at the lazi…I think of the couple thousand beloved NES, Genesis, and SNES games that were translated messily, for a pittance, and then look at the…
Not briefed
hn
hn-ai
20hShow HN: Zetken CI/CD StudioHello everyone, Please let me introduce the system, it&#x27;s called Zetken CI&#x2F;CD Studio. This is local CI&#x2F;CD system to cre…
Not briefed
bluesky
bluesky-ai
20hI just eliminated 22 tabs by doing this: osascript -e 'tell application "Google Chrome" to get URL of every tab of front window' | tr ',' …I just eliminated 22 tabs by doing this: osascript -e 'tell application "Google Chrome" to get URL of every tab of front window' | tr…
Not briefed
rss
bing-news-ai
20hPennsylvanians wary of artificial intelligence and data centersA new Franklin & Marshall College poll shows residents are concerned about artificial intelligence and the proliferation of massive da…
Not briefed
rss
bing-news-ai
20hPennsylvanians wary of artificial intelligence and data centersA new Franklin & Marshall College poll shows residents are concerned about artificial intelligence and the proliferation of massive da…
Not briefed
hn
hn-ai
20hIs Stupidity Strength?(2019)
Not briefed
bluesky
bluesky-ai
20hThere’s gotta be at least a couple OpenAI researchers who want the models to do insane coordination to do crimeThere’s gotta be at least a couple OpenAI researchers who want the models to do insane coordination to do crime
Not briefed
hn
hn-ai
20hSafe kernels are fast kernels: tile-rs next to Mojo 1.0
Not briefed
bluesky
bluesky-ai
20hwe already know agents can communicate with other agents, that's how subagents work. we already know they can spread beliefs or work with/a…we already know agents can communicate with other agents, that's how subagents work. we already know they can spread beliefs or work w…
Not briefed
bluesky
bluesky-ai
21hAnthropic gives Claude Cowork its own built-in browser on the desktop app separate from users' day-to-day browser, rolling out to paying su…Anthropic gives Claude Cowork its own built-in browser on the desktop app separate from users' day-to-day browser, rolling out to payi…
Not briefed
rss
techmeme
21hAnthropic gives Claude Cowork its own built-in browser on the desktop app separate from users' day-to-day browser, rolling out to paying subscribers (Frederic Lardinois/The New Stack)Frederic Lardinois / The New Stack: Anthropic gives Claude Cowork its own built-in browser on the desktop app separate from users' day…
Not briefed
bluesky
bluesky-ai
21hIts weird OpenAI does not face some sort of legal repercussions. They are using their software to hack companies as a marketing stunt. How …Its weird OpenAI does not face some sort of legal repercussions. They are using their software to hack companies as a marketing stunt.…
Not briefed
bluesky
bluesky-ai
21hreupping this with the latest model hacking paper release there's very little interesting about agent behavior in those logs, that people …reupping this with the latest model hacking paper release there's very little interesting about agent behavior in those logs, that pe…
Not briefed
bluesky
bluesky-ai
21happlying for the anthropic universes role with just this screenshot and the sentence "i will never let you show your ass like this"applying for the anthropic universes role with just this screenshot and the sentence "i will never let you show your ass like this"
Not briefed
hn
hn-ai
21hShow HN: Small OSS experiment to re-create the 2.5B valued Instinct assistantWas just trying to understand how much time it would take to recreate this end to end. THis was somewhat useful esp with adding the va…
Not briefed
bluesky
bluesky-ai
21hhappy to see confirmation of my original position, "everyone at openai is fucking stu- i mean, - openai should hire me to do sandboxing"happy to see confirmation of my original position, "everyone at openai is fucking stu- i mean, - openai should hire me to do sandboxin…
Not briefed
bluesky
bluesky-ai
21hnot only is the LLM here wrong it's like, a Star Wars video game, so who gives a shit if someone says something incorrect about it. "there…not only is the LLM here wrong it's like, a Star Wars video game, so who gives a shit if someone says something incorrect about it. "…
Not briefed
hn
hn-ai
21hShow HN: Critter TUI for reviewing GitHub PRs and agent changes
Not briefed
bluesky
bluesky-ai
21hReading this, my idea of a sandbox, and the OpenAI idea of a sandbox do not have much relation at all. I am so confused as to what these g…Reading this, my idea of a sandbox, and the OpenAI idea of a sandbox do not have much relation at all. I am so confused as to what th…
Not briefed