◉AnthropicBreaking
Claude Opus 4.8 sets new state-of-the-art on agentic coding evaluations
Anthropic's flagship Opus 4.8 leads on SWE-bench-style tasks with a 1M-token context option for large repositories.
ModelsJun 12, 2026
Live Intelligence
AI news & updates
Curated headlines from across the four hyperscalers. Breaking stories are flagged with a pulsing badge.
Showing 12 items · 4 breaking
◆GoogleBreaking
Gemini 2.5 Pro tops enterprise coding and long-context benchmarks
Google's latest Gemini release posts state-of-the-art results on agentic coding and 2M-token retrieval tasks.
ModelsJun 11, 2026
▲AWSBreaking
Amazon Bedrock launches unified agent runtime with cross-model handoff
Bedrock Agents can now route a single task across multiple foundation models, including Claude and Amazon Nova.
ProductJun 10, 2026
⬡MicrosoftBreaking
Azure AI Foundry adds agent orchestration for multi-step enterprise workflows
New Foundry Agent Service lets teams compose, evaluate, and govern multi-agent systems with built-in tracing and content safety.
ProductJun 9, 2026
◉Anthropic
Model Context Protocol becomes the de-facto standard for tool integration
Major IDEs and cloud platforms ship native MCP support, letting agents connect to tools and data through one open protocol.
EcosystemJun 6, 2026
◆Google
Google Agentspace brings enterprise search and agents into one workspace
Agentspace now unifies company knowledge, Gemini agents, and connectors across Drive, Confluence, and Jira.
ProductJun 5, 2026
⬡Microsoft
Copilot Studio opens autonomous agents to all Microsoft 365 tenants
Autonomous agents can now trigger from events across Teams, Outlook, and SharePoint, with admin-level guardrails.
PlatformJun 3, 2026
▲AWS
Amazon Q Developer adds repo-wide autonomous refactoring
Q Developer can now plan and execute multi-file refactors and upgrades across an entire codebase with review gates.
DeveloperJun 2, 2026
◉Anthropic
Anthropic Academy launches free enterprise AI certification track
A new structured learning path covers prompting, agents, and safe deployment, with an enterprise certificate coming soon.
TrainingMay 30, 2026
◆Google
Vertex AI adds one-click model evaluation and grounding with Google Search
New Vertex tooling lets teams benchmark candidate models and ground responses in live web results out of the box.
PlatformMay 29, 2026
⬡Microsoft
Azure OpenAI Service expands fine-tuning to new reasoning models
Enterprises can now fine-tune the latest o-series reasoning models with reinforcement fine-tuning in select regions.
ModelsMay 27, 2026
▲AWS
SageMaker unifies notebooks, pipelines, and governance in one studio
The next-gen SageMaker Studio consolidates data prep, training, and model governance for regulated industries.
PlatformMay 24, 2026