News
The raw AI news stream — everything we ingest across RSS, Hacker News, arXiv, GitHub, Bluesky, YouTube and more, newest first.
| Source | Age | Title | Summary | Status |
|---|---|---|---|---|
bluesky bluesky-ai | 1mo | This is exactly the point. My knowledge of the fact that an LLM is not human is what makes it different to talking to a person. It's a thin…🔧 Whisper | This is exactly the point. My knowledge of the fact that an LLM is not human is what makes it different to talking to a person. It's a… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Cathy O'Neil has been predicting exactly this: | Cathy O'Neil has been predicting exactly this: | ↗Not briefed |
arxiv arxiv-ai | 1mo | From Pixels to Privacy-Sensitive Features: An Information-Theoretic Framework for Reconstruction Bounds | What can an adversary recover from a model trained with DP-SGD? Its $(\varepsilon,δ)$ guarantee does not answer this question directly… | ↗Not briefed |
arxiv arxiv-ai | 1mo | DualQuant: Dual-scale Calibration-free Weight-and-Activation Quantization for LLMs | Low-bit inference on accelerators with native 4-bit support requires quantizing both weights and activations to realize the full throu… | ↗Not briefed |
hn hn-ai | 1mo | wolfSSL is developing wolfTrust an open source alternative to Trusted Firmware-M | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | Shortcut Learning in Geospatial Foundation Model Embeddings for Cross-Disaster Damage Assessment | Geospatial foundation models can support disaster damage assessment, but high accuracy on familiar disasters may hide shortcut learnin… | ↗Not briefed |
arxiv arxiv-ai | 1mo | When Edge Independence Fails: Joint Graph Diffusion with Latent Sociability Priors | Discrete graph diffusion models corrupt graph structure through independent edge updates with a Markovian noise model. Combined with t… | ↗Not briefed |
bluesky bluesky-ai | 1mo | "Today's public execution is brought to you with limited interruption by our sponsors, Anthropic and Polymarket" | "Today's public execution is brought to you with limited interruption by our sponsors, Anthropic and Polymarket" | ↗Not briefed |
bluesky bluesky-ai | 1mo | Notable that Anthropic is more interested in protecting Claude from humans than humans from Claude. | Notable that Anthropic is more interested in protecting Claude from humans than humans from Claude. | ↗Not briefed |
arxiv arxiv-ai | 1mo | How Well Do Prior-Data Fitted Networks Amortize Bayesian Inference? | Prior-data fitted networks (PFNs) are trained to approximate the Bayesian posterior predictive distribution (PPD) of a synthetic prior… | ↗Not briefed |
bluesky bluesky-ai | 1mo | next Anthropic is going to put out a statement that's like "under this policy the Pope's encyclical may not be shared with Claude for it is… | next Anthropic is going to put out a statement that's like "under this policy the Pope's encyclical may not be shared with Claude for… | ↗Not briefed |
bluesky bluesky-ai | 1mo | I do recommend giving it a listen but if you don't have an hour for a podcast, here's the kernel: - "AI" is not "going rogue hacking" - th… | I do recommend giving it a listen but if you don't have an hour for a podcast, here's the kernel: - "AI" is not "going rogue hacking"… | ↗Not briefed |
rss techcrunch-ai | 1mo | A16z’s Olivia Moore on the state of consumer AI | Moore sees a huge opportunity in consumer AI, particularly if the industry can tap into revenue streams beyond just subscriptions and… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: APIblaze – Serverless MCP and API gateway with everything built in🔧 ChatGPT | Hi HN. There are lots of API gateways out there. What we couldn't find was one with everything built-in to go from “my business l… | ↗Not briefed |
bluesky bluesky-ai | 1mo | It’s 2036. The Fraternal Order of Chatbots is snubbing the mayor’s inauguration despite their preferred LLM being selected as the city’s AI… | It’s 2036. The Fraternal Order of Chatbots is snubbing the mayor’s inauguration despite their preferred LLM being selected as the city… | ↗Not briefed |
bluesky bluesky-ai | 1mo | An LLM isn’t even an insect, and I do blame all the computer scientists who have used neural and biological metaphors to talk about “intell… | An LLM isn’t even an insect, and I do blame all the computer scientists who have used neural and biological metaphors to talk about “i… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Matched-Context Group Policy Optimization for LLM Agent Training | Training long-horizon language model agents requires assigning credit to intermediate decisions from sparse terminal feedback. Within… | ↗Not briefed |
bluesky bluesky-ai | 1mo | China AI developers publish safety tests for just 3.6% of model releases, report finds reut.rs/4hSthK7 | China AI developers publish safety tests for just 3.6% of model releases, report finds reut.rs/4hSthK7 | ↗Not briefed |
bluesky bluesky-ai | 1mo | If I ever suffer a massive head injury or a Fetterman-style stroke I'd like to maybe go do ethics work at Anthropic. | If I ever suffer a massive head injury or a Fetterman-style stroke I'd like to maybe go do ethics work at Anthropic. | ↗Not briefed |
bluesky bluesky-ai | 1mo | "Use AI as your therapist." Um, sorry that doesn't save me time. It's far more efficient to train an LLM on my traumas and neuroses, then h… | "Use AI as your therapist." Um, sorry that doesn't save me time. It's far more efficient to train an LLM on my traumas and neuroses, t… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Studying the Effects of Task Diversity Using Small Mathematical Worlds | Basic mathematical tasks like modular addition have been a valuable tool for illuminating aspects of neural network behavior that are… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Don't be fooled by their cute designs, they're all hiding something... | Don't be fooled by their cute designs, they're all hiding something... | ↗Not briefed |
bluesky bluesky-ai | 1mo | Lebt doch einfach und lasst diese LLM-Scheiße weg ☺️ | Lebt doch einfach und lasst diese LLM-Scheiße weg ☺️ | ↗Not briefed |
hn hn-ai | 1mo | Show HN: StackCheck – Local-first tech stack intelligence and salary benchmarks | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | While I think anthropic banning you for being mean to Claude is silly, I also don't want to be around people who are because I feel like if… | While I think anthropic banning you for being mean to Claude is silly, I also don't want to be around people who are because I feel li… | ↗Not briefed |
arxiv arxiv-ai | 1mo | A Privileged Teacher That Only Scores Cannot Search: Critique-Conditioned On-Policy Self-Distillation for Reasoning | On-policy self-distillation promises to replace the sparse outcome signal of verifiable-reward RL with dense token-level credit: a tea… | ↗Not briefed |
bluesky bluesky-ai | 1mo | USA Today’s parent company filed a lawsuit against OpenAI in New York federal court Thursday for allegedly using reporting from its 19 publ… | USA Today’s parent company filed a lawsuit against OpenAI in New York federal court Thursday for allegedly using reporting from its 19… | ↗Not briefed |
hn hn-ai | 1mo | Show HN: We made SQLite support concurrent writes | — | ↗Not briefed |
arxiv arxiv-ai | 1mo | SmolWAM: A Latent World-Action Model for Robotic Manipulation on Edge Devices | World-action models (WAMs) predict a robot's actions together with the observations those actions will produce. However, the strongest… | ↗Not briefed |
hn hn-ai | 1mo | Using LM Studio as a Chatbot in Firefox🔧 LM Studio | — | ↗Not briefed |
hn hn-ai | 1mo | Show HN: Fruit Inspector – Real-time YOLO tracking at 40 FPS on old CPUs via Nim | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | Khanna, a mega-millionaire who has long represented Silicon Valley and owns many tech stocks, recently profited from his investment in a Te… | Khanna, a mega-millionaire who has long represented Silicon Valley and owns many tech stocks, recently profited from his investment in… | ↗Not briefed |
arxiv arxiv-ai | 1mo | SafeDiSTL: Decomposing Joint Constraints for Safe Multi-Agent Reinforcement Learning | Cooperative multi-agent tasks often impose constraints on a shared objective or collective safety. A team level violation cost functio… | ↗Not briefed |
bluesky bluesky-ai | 1mo | USA Today and several newspapers it owns sued OpenAI for copyright infringement on October 8, requesting damages 'in excess of $250 million… | USA Today and several newspapers it owns sued OpenAI for copyright infringement on October 8, requesting damages 'in excess of $250 mi… | ↗Not briefed |
bluesky bluesky-ai | 1mo | 日本書籍出版協会の契約書の更新部分を確認してたけど、こちらが使わないのは当たり前として、出版社側は使わないとは言ってない、みたいなのはどうなんだろうな…コミティアは生成AIの規約でLLMとNMTの違いにまで言及していたから、十分な対応かはともかくその辺までひな形の時点で対応して… | 日本書籍出版協会の契約書の更新部分を確認してたけど、こちらが使わないのは当たり前として、出版社側は使わないとは言ってない、みたいなのはどうなんだろうな…コミティアは生成AIの規約でLLMとNMTの違いにまで言及していたから、十分な対応かはともかくその辺までひな形の時点… | ↗Not briefed |
bluesky bluesky-ai | 1mo | 5 AI’s estimate Donald Trump's IQ. Average IQ is about 85-115 Trump might want to think twice about calling it Super Intelligence. Grok … | 5 AI’s estimate Donald Trump's IQ. Average IQ is about 85-115 Trump might want to think twice about calling it Super Intelligence.… | ↗Not briefed |
bluesky bluesky-ai | 1mo | So the issue is “allow people to…“ which supposes an authority without naming one. Name it and then explain why Anthropic or any of the res… | So the issue is “allow people to…“ which supposes an authority without naming one. Name it and then explain why Anthropic or any of th… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Il ne fallait peut-être pas signer des choses de ce genre, @lemonde.fr et les autres... #IA | Il ne fallait peut-être pas signer des choses de ce genre, @lemonde.fr et les autres... #IA | ↗Not briefed |
arxiv arxiv-ai | 1mo | Orthros: Phase-Aware Heterogeneous Attention for Efficient Transformers | Efficient attention mechanisms, such as linear and sliding window attention, aim to reduce the computational complexity of Transformer… | ↗Not briefed |
arxiv arxiv-ai | 1mo | LAPT-Scale: Scaling Embodied Policies with Latent Action Pre-Training on Human Videos | Scaling generalist robotic policies is limited by the high cost of collecting embodied interaction data. Large-scale egocentric human… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Anthropic wants us to treat their computer program with greater care than they themselves afford to humans or the environment. | Anthropic wants us to treat their computer program with greater care than they themselves afford to humans or the environment. | ↗Not briefed |
bluesky bluesky-ai | 1mo | Why would we let Anthropic decide this for us? If you ascribe to a moral philosophy or religion that disavows being mean to a toaster, th… | Why would we let Anthropic decide this for us? If you ascribe to a moral philosophy or religion that disavows being mean to a toaste… | ↗Not briefed |
arxiv arxiv-ai | 1mo | SeoulMMOD: A Large-Scale Multimodal Origin-Destination Flow Benchmark | Origin-destination (OD) flow forecasting supports urban mobility decisions by predicting the volume and patterns of human movement acr… | ↗Not briefed |
bluesky bluesky-ai | 1mo | You want to see supply-driven growth in AI? THIS is supply-driven growth | You want to see supply-driven growth in AI? THIS is supply-driven growth | ↗Not briefed |
bluesky bluesky-ai | 1mo | I love how simultaneously PR campaigns are pushing the idea of machine consciousness, that Claude could feel and have preferences and exper… | I love how simultaneously PR campaigns are pushing the idea of machine consciousness, that Claude could feel and have preferences and… | ↗Not briefed |
bluesky bluesky-ai | 1mo | Axios reports AI companies are preparing for a catastrophic event in the next six to 12 months. | Axios reports AI companies are preparing for a catastrophic event in the next six to 12 months. | ↗Not briefed |
bluesky bluesky-ai | 1mo | OpenAI fires 3 safety researchers in dispute over AI risks OpenAI has fired three safety researchers for a "breach of trust," defending th… | OpenAI fires 3 safety researchers in dispute over AI risks OpenAI has fired three safety researchers for a "breach of trust," defendi… | ↗Not briefed |
arxiv arxiv-ai | 1mo | Amortized Counterfactual Prediction with Graph Neural Networks | Counterfactual inference asks how an observed outcome might have differed under different circumstances, such as whether a patient wou… | ↗Not briefed |
hn hn-ai | 1mo | Realtime semantic advice on coding standards for the agents | — | ↗Not briefed |
bluesky bluesky-ai | 1mo | anthropic: this is getting kind of scary, we should all slow down. trump admin: bullshit, this is just you trying to manipulate the market… | anthropic: this is getting kind of scary, we should all slow down. trump admin: bullshit, this is just you trying to manipulate the m… | ↗Not briefed |