AI
Artificial intelligence: LLMs, agents, diffusion, vision, NLP and the latest from top labs.
593 published articles
AI governance
China's tech giants just deleted millions of AI relationships
ByteDance, Tencent, and Alibaba are simultaneously shutting down AI companion features that allowed millions of users to form emotional bonds with customizable virtual characters. The move, widely seen as a response to China's new AI content regulations taking effect in April 2026, has triggered user outrage and complaints about losing irreplaceable emotional investments.
2026-07-08
Agent Frameworks
IBM's new open-source agent framework cuts the boilerplate and keeps the brains
IBM's open-source CUGA framework flips the typical agent development model on its head by handling orchestration, state management, and planning. Developers are left to write only a tool list and a prompt. Over two dozen single-file apps demonstrate the approach, from a movie recommender to a multi-agent lead-generation system, all deployable in governed production without needing a rewrite.
2026-07-07
Artificial Intelligence
Ifbench reveals the instruction-following gap that other benchmarks miss
IFBench measures language models' ability to follow precise natural-language instructions. xAI's Grok models lead while Claude models lag, showing instruction following is a distinct capability from general intelligence.
2026-07-07
Generative AI
The AI content factory is erasing the value of your byline
Synthetic content tools are churning out vast quantities of articles, images, and video, threatening to undermine trust in digital authorship and the economic value of human-created work. The efficiency gain comes at a steep cost: blurred lines between real and generated.
2026-07-07
International expansion
Anthropic planted a flag in Seoul. The real story is who showed up.
Anthropic opens a Seoul office and announces partnerships with NAVER, Nexon, LG CNS, Hanwha Solutions, and Samsung SDS. An MOU with Korea's Ministry of Science and ICT focuses on AI safety and cybersecurity. The real story is a market that treats safety as a feature, not a cost.
2026-07-07
Efficient World Models
Fast-LeWM just made visual planning stop stumbling over its own steps
Researchers introduce Fast-LeWM, a latent world model that speeds visual planning by predicting future states from action prefixes in parallel. The approach cuts computational costs and error buildup, outperforming prior one-step transition models.
2026-07-07
Open Source & Local AI
Local LLMs just ate cloud triage for lunch
A maintainer's deep dive into replacing costly cloud-based triage with local open-weight models on an NVIDIA GB10 reveals that models like Gemma 4 and Qwen 3.6 can achieve F1 scores above 0.80 on real-world issue classification. The system uses an agent harness, a restricted read-only shell, and structured outputs to deliver near-instant notifications with zero API spend.
2026-07-07
AI Regulation
Export controls lifted, Claude Fable 5 returns with a jailbreak fix that mostly works
Anthropic redeploys Claude Fable 5 and Mythos 5 after US export controls are lifted. The company outlines safeguard updates, a proposed jailbreak severity framework, and new commitments to government collaboration on AI security.
2026-07-07
Enterprise AI partnership
TCS is betting 50,000 employees on Claude. That's the easy part.
Tata Consultancy Services partners with Anthropic to deploy Claude to 50,000 employees and create industry-specific AI offerings for regulated sectors. The deal strengthens Anthropic's presence in India, its second-largest market, and signals a deeper enterprise push.
2026-07-07
Paper 2606.23050 decoded
A 33-page preprint just landed. Here's what it reveals about where AI is heading
A 33-page preprint, paper 2606.23050, released only five days ago, marks a serious contribution to AI research. This article unpacks its methodology, key findings, and what it signals for the trajectory of machine learning and natural language processing.
2026-07-07
Artificial Intelligence
Ai2's EMO: the MoE model where modularity emerges from data, not human priors
Ai2's new MoE model, EMO, uses a novel training method that lets expert modules emerge naturally from data, enabling selective expert use with minimal performance loss. The model matches standard MoE performance on benchmarks while offering vastly improved modularity.
2026-07-07
Editorial opinion
The prompt to "generate an article" is the problem, not the solution
A blank prompt to "generate an article" reveals a fundamental misunderstanding of journalism. The hardest work is not writing, but deciding what is worth writing about, a distinction that machines still cannot make.
2026-07-07