Google Research is turning enterprise RAG into an iterative agent workflow, not a one-shot retrieval step. Its sufficient-context check lifted factuality accuracy by up to 34% and reached 90.1% accuracy in a cross-corpus FramesQA setup.
OpenAI frontier models and Codex are now generally available on Amazon Bedrock. The post drew more than 1.2M views and points to a broader AWS path for enterprise AI, including future Daybreak security capabilities.
NVIDIA is packaging a 550B-parameter MoE model with agent tooling instead of treating the model as a standalone release. The pitch is concrete: up to 5x faster inference, up to 30% lower cost, and availability beginning June 4.
A Daniel Miessler post says Claude Code is preparing a /workflows feature, drawing more than 269K views. The signal is a shift from one-off coding prompts toward repeatable SOP execution inside enterprise AI systems.
HashiCorp founder Mitchell Hashimoto's warning about companies in 'AI psychosis' hit 1,650 points on Hacker News. His thesis: AI's ability to fix bugs quickly is masking accumulating systemic risk that no one can see in local metrics.
OpenAI has launched the OpenAI Deployment Company (DeployCo), a majority-controlled subsidiary backed by $4B from 19 investors to help enterprises deploy frontier AI at scale.
Cloudflare reported a 600% surge in AI usage in Q1 2026 while simultaneously announcing layoffs of 1,100 employees (20% of workforce) as agentic AI 'fundamentally changes' the company's operations.
On its Q1 2026 earnings call (April 29), Meta disclosed that its business AI tools processed about 10 million conversations per week as of late March, up from 1 million at the start of the year — still free, monetization hinted ahead.
Sierra, co-founded by Bret Taylor, closed a $950M Series E at $15.8B valuation on May 4. The enterprise AI agent company has over 40% of the Fortune 50 as customers and $150M ARR.
Microsoft’s AI push is no longer a capex story waiting for payoff. In results released on April 29, the company said its AI business surpassed a $37 billion annual revenue run rate, while Azure and other cloud services revenue grew 40% and commercial remaining performance obligation jumped 99% to $627 billion.
LocalLLaMA paid attention to Granite 4.1 because IBM went in the opposite direction from giant reasoning hype: a broad release built around dense 3B, 8B, and 30B language models tuned for instruction following and tool calling. Comments welcomed the extra competition, but also pushed back on how strong the benchmarks really are.
Why it matters: most agent stacks look solid in a notebook and fragile in production. Mistral says Workflows is already running critical processes at six named organizations and adds the pause-resume, observability, and fault tolerance teams usually have to build themselves.