A parameter selection technique has been proposed that reduces forgetting of existing knowledge while training only on new data.
As AI hyperscalers' borrowing surges, warning signs are flashing over the sustainability of their financing.
Meta doubled the training efficiency of its ads recommendation foundation model GEM while scaling training compute 4x.
Vercel WAF for Blob is now generally available on all plans.
Google Cloud presented a phased mainframe modernization strategy leveraging AI.
MirrorCode evaluated long-horizon autonomous coding ability and revealed Claude Opus 4.7's results.
AWS cut on-demand inference pricing for OpenAI's GPT-5.6 models on Bedrock by up to 80%.
Microsoft has released Orchard, an open-source framework for scaling agent training and evaluation.
Stripe built Kai, a company-wide AI agent, in just one week using Deep Agents.
An overview of leading use cases and adoption requirements for conversational AI that automates HR work from recruitment to offboarding.
Apple TV will live stream all 62 matches of the Leagues Cup, which kicks off on August 4, in more than 100 countries.
Kakao integrated LM-SPT into Kanana-o, improving voice efficiency and style control, achieving 94.50 on the Korean InstructTTSEval benchmark, close to Gemini-2.5-flash-preview-tts's 95.38.
353,000 people registered for Kaggle and Google's AI agent vibe coding course.
A German court ruled that Suno's unauthorized use of copyrighted works constitutes copyright infringement.
Cloudflare has unveiled a more scalable computer runtime for agents than containers.
Cloudflare Workers RPC now supports calls between Python and JavaScript.
Cloudflare reduced GPU memory usage and increased throughput for Kimi and GLM.
Cloudflare has launched the Billable Usage API, which lets you look up per-product usage and costs by account.
Cloudflare has added inbound TCP connections and gRPC execution capabilities to Workers and Containers.
ByteDance is integrating Seedance into Gauth to provide AI animated lessons.
A SQLite vulnerability report that appears to have been AI-generated was registered as a CVE despite no actual flaw existing.
AirLLM has revealed a memory-saving method that runs a 70B model on a 4GB GPU.
AI9Stars has released a 39B open-weight language model based on Apache 2.0.
SimpleMem compresses long-term memory to boost LoCoMo performance while cutting inference tokens by up to 30x.
AI Briefing