The month's main changes in models, products, applications, and developer tools. Each item links to our full story and the original source.
Google introduced Gemini 4 Argon, featuring a 1 million output token limit. Trusted defenders currently access it without cyber guardrails via the Fairwind Program.
The increased output capacity allows the model to sustain deep reasoning across complex, long-horizon professional workflows.
OpenAI has introduced GPT-6.1 Sol, a cost-efficient model providing performance comparable to GPT-6 Astra for coding and professional tasks.
The release provides a more economically viable option for developers building complex AI agents and software engineering tools that require high reasoning capabilities.
Anthropic positions Claude Sonnet 5.5 for everyday work. Here is what changed and why it may matter for AI teams.
For teams running a high volume of routine tasks, speed and cost can change the economics of automation. Those figures come from the model provider, so teams should compare results on their own workloads before switching.
OpenAI has introduced GPT-6 Sol and Luna, featuring reduced API costs and enhanced prompt caching for professional and enterprise use.
These improvements address the cost-efficiency of integrating advanced AI into scalable workflows, allowing for more economical handling of context-heavy applications.
The platform provides predictions for 9 billion single-nucleotide variants to assist researchers in interpreting human genome data.
AlphaGenome Atlas enables researchers to identify and prioritize potential causal variants in rare disease studies and population genetics, accelerating genomic discovery.
OpenAI has introduced GPT-6 Astra, a model achieving top marks on complex reasoning benchmarks and improved autonomous computer operation.
This release marks a shift toward highly autonomous technical workflows, highlighting the dual-use nature of models capable of both vulnerability discovery and defense.
Google has launched Gemini 3.8 Flash for agentic reasoning and a specialized Cyber version for automated vulnerability detection and patching.
These models provide enterprises and developers with cost-efficient tools for autonomous agent workflows and automated software security patching.
OpenAI is rolling out Dots, persistent AI agents that manage third-party applications and perform proactive research for Pro and Enterprise users.
The launch of Dots shifts AI utility from interactive chat to persistent automation. By offloading multi-step tasks to agents that operate 24/7, users can minimize manual coordination between different software platforms.
Microsoft has upgraded Copilot to include autonomous task management, custom application building, and a new usage-based billing model.
This update shifts the product from a static information retrieval tool to an active agent that handles complex workflows and software development. For businesses, this enables the automation of recurring tasks and direct integration of AI-driven solutions into existing enterprise environments.
Salesforce unveiled AIforce, an interface layer that integrates CRM data and logic into tools like Slack and Claude with a zero-data-retention policy.
The platform streamlines workflows by bringing CRM logic into collaborative environments, reducing administrative friction and context switching.
OpenAI launches a specialized ChatGPT workspace featuring GPT-6 Astra and integrated premium financial data for research and modeling.
By integrating high-quality financial data with reasoning capabilities, the platform aims to streamline the creation of research reports and financial models while maintaining enterprise security standards.
Meta has introduced new skills and connectors to its Muse AI agent to assist small business owners with operational automation.
By connecting scattered administrative tasks and communications into a single AI-driven interface, Meta aims to help small business owners save time on manual operations.
Google Vids now features Gemini Omni 1.1 Flash, enabling 1080p video generation, clip upscaling, and precise scene duration controls for all users.
This update improves video creation by providing granular control over timing and visual consistency, helping users align generated content precisely with their narrative and voiceover needs.
Google uses its Mantis multi-agent framework to scan code pre-submit and autonomously patch vulnerabilities, preventing hundreds of issues monthly.
This approach shifts security earlier into the development lifecycle, allowing for continuous, low-latency vulnerability detection and automated resolution without manual intervention.
Basis, Clay, and Exa Labs utilize AI agents for onboarding, account management, and developer tasks with mandatory human-in-the-loop review points.
The examples suggest a practical pattern: define a repeatable task, give the agent relevant context, and keep a person responsible for decisions. The reported time savings come from the companies involved.
NVIDIA has introduced the Open Agent Safety Platform, designed to secure AI agents through hardware-enforced runtime monitoring and policy management.
The platform addresses the risks posed by autonomous AI agents by moving security beyond the application layer. By utilizing hardware-level enforcement, organizations can maintain control over agent behavior and prevent them from bypassing standard security policies.
Google has updated its Private AI Compute platform with secure, persistent storage that keeps encryption keys on user devices to maintain context privately.
This shift moves beyond the previous stateless architecture, which discarded context after each task. It allows for personalized AI interactions across multiple devices without requiring users to sacrifice data privacy or security.
GPT-6 now supports improved prompt caching with up to 90% discounts on input tokens and flexible reasoning adjustments without breaking cache.
This update reduces operational costs and latency for persistent AI agents. The ability to modify reasoning effort while maintaining cache efficiency allows for more granular control over model behavior.
AWS has introduced the AgentCore runtime, featuring snapshot-based startup optimization and improved memory reclamation to enhance agent performance.
This update addresses startup bottlenecks and offers potential cost reductions by tracking memory usage dynamically rather than billing based on peak watermarks.
Google Cloud introduces Agent Substrate on GKE, an open-source runtime designed to improve density and efficiency for autonomous AI agent workloads.
For organizations running autonomous agent systems on Kubernetes, managing memory and CPU footprints for idle instances is a primary hurdle. This runtime addresses the challenge by decoupling agent state from compute resources, enabling higher density without compromising security.
OpenAI's GPT-Live-1 API allows developers to build full-duplex voice agents capable of simultaneous listening and speaking with native handling of interruptions.
This API allows developers to engineer responsive voice agents and phone workflows that maintain a natural conversational flow by handling interruptions and pauses in real-time.
OpenAI launched the Agents API public beta, featuring durable sessions, context compaction, and managed infrastructure for long-running AI agent execution.
This API centralizes session orchestration and execution, enabling developers to build persistent agents capable of handling complex, long-term workflows.