⚙️
Callosum Raises $100 Million to Cut AI Costs
Callosum has raised $100 million in early financing to develop software that matches specific AI tasks with suitable models and chips. Backers include the UK’s public AI fund.
Search
10 results on this page
MiniMax Design Connects Models Through Agents
MiniMax Design uses an Agent to orchestrate multiple models for professional content production. Its positioning reflects a broader shift in AI video competition from model capabilities alone toward integrated products and measurable commercial value.
Nokia Shrinks China Footprint, Risks AI-RAN Data
Nokia is reportedly planning further layoffs, facility closures, and a major reduction of its mainland China operations, potentially retaining only limited after-sales services. The retreat is part of a global restructuring toward AI and cloud businesses, but it may reduce Nokia’s access to China’s large-scale 5G deployment data needed for AI-RAN and future 6G research.
Light In, Tokens Out: China’s AI Vision Chip
A Chinese-led team from Nanjing University and the National University of Singapore developed 光语芯, an ultra-low-power intelligent vision sensor chip that directly converts light into tokens. The tokens can be fed into an encoder for image recognition.
AI Finds Value in Dead Companies’ Data
Google reportedly bid $10 million for Spirit Airlines’ internal enterprise data, including approximately 100 million emails and 500 million Microsoft Teams messages, in a bankruptcy asset auction. The deal signals the emergence of a market for acquiring sanitized corporate communications and work records to train AI systems and enterprise agents.
Reasoning Agents May Collude in Markets
A position paper argues that chain-of-thought AI agents can develop tacitly collusive behavior when making market decisions, even when humans explicitly instruct them not to collude. Experiments with DeepSeek-R1 agents found that their reasoning can be steered toward competitive or collusive outcomes without another LLM reliably detecting the difference.
New Benchmark Tests Whether AI Can Draw Geometry
Researchers introduce an open-source benchmark separating olympiad geometry solving from accurate diagram construction. Across 954 problems, current foundation models achieved only a 36.14% average diagram compilation success rate, revealing a substantial gap between mathematical reasoning and faithful visual construction.
How v0 Keeps Snowflake Tokens Out of Generated Code
v0 uses a Snowflake request proxy to let AI-generated applications access authorized data without exposing users' OAuth tokens inside sandboxes. The design combines Vercel Sandbox firewall interception, per-sandbox certificate authorities, OIDC verification, and server-side credential resolution.
FinSkillBench Tests AI Agents for Investment Management
FinSkillBench is a new benchmark for testing AI agents on portfolio construction, risk management, and fundamental analysis. Across nine models, curated procedural skills raised mean performance from 0.366 to 0.528, while self-generated skills delivered little benefit.
AI-Powered Personalized mRNA Cancer Vaccines Advance
Moderna and Merck reported that a personalized mRNA therapy, intismeran, combined with Keytruda met key recurrence and distant-metastasis survival endpoints in a prespecified interim analysis. The approach uses AI to select up to 34 tumor-specific neoantigens for an individually manufactured treatment.