AI tools explained for people who use them.

Anthropic opened the Claude Marketplace on March 6, consolidating procurement for six partner tools under a single Anthropic billing relationship—streamlining enterprise software acquisition. Multi-cloud availability expanded: all Claude models now run on Anthropic API, AWS Bedrock, and Google Vertex AI, eliminating vendor lock-in for large organizations. This move addresses a critical enterprise need: infrastructure portability. Companies cannot migrate entire stacks to single-vendor platforms. By supporting AWS and Google infra alongside Anthropic, Claude becomes the AI layer across organizations already invested in multi-cloud strategies. Regional processing options (with 10% premium for data residency) further appeal to regulated industries (finance, healthcare, government).

The new frontier model includes 1.05M token context window and three variants: Standard, Thinking, and Pro. The Thinking variant competes directly with other reasoning models. 33% fewer factual errors than GPT-5.2. Available to Plus subscribers immediately with new Tool Search architecture enabling more intelligent dynamic tool calling.

Anthropic releases Claude Sonnet 4.6 with 1M context window in beta, at unchanged $3/$15 pricing. Independent testing shows it's preferred over Opus 4.5 in Claude Code 59% of the time for complex development tasks. Represents significant capability jump in free tier.

Chinese AI lab releases 1-trillion-parameter model with open weights and 1M token context. Multimodal support for text, code, vision, and audio. Unlike US models, DeepSeek comes with no investment fees, no classified obligations, and no geopolitical restrictions. Strategic victory for Global South AI access.

Analysis of March 2026 shows unprecedented acceleration in AI development cadence. US labs, Chinese competitors, and open-source community are each racing releases. March 3-15 alone saw GPT-5.4, DeepSeek V4, Claude Sonnet 4.6, Qwen 3.5, and 10+ other major releases. Experts warn this pace makes it harder for enterprises to standardize.

Alibaba released Qwen 3.5 (0.8B, 2B, 4B, 9B parameters) on March 1, 2026 ahead of its own DeepSeek conference. Timing signals Alibaba's urgency to show competitive progress. Open-source models are increasingly practical for fine-tuning and on-device deployment.

Baidu targets the biggest blocker in agent adoption: environment setup friction. Ducclaw runs agents directly in browser with no configuration, servers, or API key management. Removes the barrier that kills most AI agent experiments before they start.

Codex reenters top-5 dev tools with parallel sandboxed execution, deep GitHub integration, and automatic PR creation. The strongest pick for teams already in the OpenAI ecosystem. Competes directly with Claude Code and JetBrains AI.

Portfolio company of 1752vc releases AI-driven analytics platform to help enterprises turn complex data into reliable insights. Targets non-technical users who lack SQL expertise. Funding enables rapid enterprise sales expansion.

New security capability gives real-time visibility into which employees use unsanctioned AI tools like ChatGPT and Gemini. Launched March 17, 2026. Major enterprise security update addressing the growing shadow AI risk in organizations.

Desktop application for managing Model Context Protocol (MCP) configurations. Runs entirely locally, no account required, no paid tiers. Solves the fragmentation problem of MCP servers across ChatGPT, Claude, and other AI platforms.

On March 9, 2026, Microsoft announced Anthropic's Claude Sonnet models are now available to M365 Copilot users. Gives enterprise customers choice between OpenAI GPT and Anthropic Claude within Microsoft Office suite. Major win for Anthropic's enterprise strategy.

Tech veteran KP Reddy launches AI-native platform for construction, backed by General Catalyst. Targets $200B+ RFI market where information delays cause project overruns. Early traction suggests enterprise AI is moving beyond software into physical work coordination.

Wealth management startup unveils AI assistant designed to streamline wealth advisor workflows. Launched March 16, 2026 at T3 Tech Conference. Targets the $200B+ advisor tech market and aims to enhance rather than replace human advisors.

On March 5, 2026, Salesforce announced six agentic AI tools for medical professionals. Handles referrals, identifies disease patterns, manages insurance claims, tracks coverage, and automates prior authorization. Reflects healthcare's urgent need for administrative automation.

Researchers publish framework for AI assistants that run entirely on-device. Addresses privacy concerns and deployment costs of cloud-based models. Practical for regulated industries, healthcare, and organizations with data residency requirements.

AI models now capable of analyzing massive datasets from unstructured sources. Google demonstrated Gemini categorizing and mapping flood events across decades of news. Shows emerging use case: AI for environmental monitoring and disaster tracking at scale.

OpenAI's Thinking variant places at or above human expert level on economically valuable tasks according to GDPVal benchmark. Morgan Stanley warns this is the beginning of "transformative AI" that will reshape labor markets and economics in 2026.

Investment bank warns major AI labs are accumulating compute for a "shock" improvement in capabilities. Executives predict workforce reductions from AI efficiency gains. OpenAI CEO suggests 1-5 person companies could soon outcompete large incumbents. Signals AI disruption is accelerating.

At Enterprise Connect 2026, Zoom revealed new agentic features to turn Zoom from meeting platform into autonomous work orchestrator. Can summarize discussions, execute actions, and coordinate across apps. Reflects shift from AI copilots to fully agentic enterprise tools.