From the archive
Follow the latest AI news and artificial intelligence developments across models, agents, robotics, research and the wider industry. Explore new AI tools, advances in computing, business moves, safety and policy, with clear reporting and useful sources. Read the stories, catch up with News in Brief and join the conversation.
Lead story
Anthropic has launched the Claude Frontier Academy, committing $100 million to train 10,000 Frontier Deployed Engineers by 2028. The programme starts with Accenture and Bain, putting a substantial workforce plan behind the push to deploy AI in organisations.
Read the report & discussion ↗Watch Desk
GitLab’s Threat Research Group says a flaw in DeepSeek-Reasonix Studio could let a hostile repository run attacker-controlled code when a developer views a file diff. GitLab says fixes are available, and the report points to a wider risk for coding tools that invoke Git on repositories they do…
Watch Desk
Conway Research is building Underdog, a personal AI product intended to run on devices people own, including offline. Andreessen Horowitz says it is leading the company’s first funding round, though the announcement gives no amount.
Cerebras Watch
General Compute has signed a multi-year agreement to deploy Cerebras wafer-scale AI hardware and offer access through its cloud platform from the first quarter of 2027. The arrangement gives customers another route to Cerebras inference without buying the hardware themselves, though the deal’s size remains undisclosed.
Databricks Watch
Databricks has outlined a Genie One workflow in which AI agents ingest custodian files, reconcile positions and prepare a preliminary net asset value before the working day starts. The important boundary: the valuation team reviews that figure, rather than leaving agents to settle what investors’ money is worth.…
OpenAI Watch
OpenAI has launched GPT-6.1 Sol at $2 per million input tokens and $10 per million output tokens, with cached input priced at $0.10. That makes the cost of using a frontier model, especially for repeated or agentic workloads, the story to watch.
OpenAI Watch
SpecterOps plans to launch an eight-hour course on securing large language models and AI agents, developed with OpenAI. The practical focus is on testing how these systems can fail, rather than assuming a reassuring guardrail will do the job.
Watch Desk
AI cloud provider Lambda has secured a loan worth more than $1bn to buy over 30,000 Nvidia GPUs, according to The Information. The fixed-rate facility carries a 6.78% interest rate, putting a concrete price on one route to financing the AI computing boom.
Watch Desk
An AI system called Ataraxos beat top-ranked Stratego player Pim Niemeijer 15 games to one, with four draws, in a 20-game series. The result, reported by Ars Technica, is notable not just for the win: the researchers say they trained the system for less than $8,000 using 16…
Watch Desk
High-yield bonds have financed at least 20 data-centre projects over the past year, and some projects serve companies building AI infrastructure. The Information says that debt is appearing not only in specialist AI funds but in broader investment portfolios too.
Watch Desk
Business Insider reports that robotics start-up FieldAI has raised $700 million at a $10 billion valuation, up from $2 billion in 2025. The reported round is a striking bet on a company building a general-purpose AI system for robots, from humanoids to drones.
Microsoft AI Watch
Microsoft’s 2026 Digital Defense Report recommends that public bodies tightly control what AI agents can access, remember and do. The practical message is to give agents narrow, temporary permissions, and keep consequential actions behind review.
AWS AI Watch
AWS says fine-tuning a Qwen3.6-27B search agent with multi-turn reinforcement learning improved its results on three of four held-out benchmarks, while sharply reducing failed tasks on BrowseComp-Plus. The useful detail is that the training rewarded the whole search journey, not just one answer at a time.
Watch Desk
Researchers say a code-execution approach helped robot agents complete more simulated tasks while using fewer language-model resources. In 700 task instances, PyRUA-Lean raised reported success from 63.1% to 71.7% under equal LLM-call budgets, a result that could matter for building more efficient robot agents.
Microsoft AI Watch
Microsoft’s Agent Framework 1.20 adds native computer-use support, new vector-store connectors and changes to how Foundry hosts agent workflows. For developers, the headline is a broader set of building blocks, with some breaking changes to check before upgrading.
Watch Desk
Context Language Models treat a model’s working context as an editable file, which it can change as a task unfolds. The researchers say their approach outperformed existing context-management strategies across several tasks while using fewer floating-point operations.
Cohere Watch
Cohere has released Embed 5, a pair of embedding models for finding information in enterprise documents, images and other data. Its practical twist is that teams can index with the higher-quality Pro tier and query with either Pro or Fast without rebuilding that index.
Allen Institute for AI Watch
Ai2 has open-sourced AstaBrief, an 8-billion-parameter model for turning a research question and retrieved literature into a cited report. Ai2 says its Fast mode averaged 51.1 seconds per report, compared with 178.5 seconds for the Claude-powered Thinking mode it tested.
