Snowflake and Anthropic sign USD $200m AI agents deal

In partnership with

The Latest Agentic Development

Databricks Debuts Instructed Retriever for Agent Accuracy Boost

Databricks released Instructed Retriever, a new data retrieval architecture delivering 70% better accuracy than traditional RAG systems and 30% improvement in multi-step agentic processes with 8% fewer steps. Available today in beta through its Knowledge Assistant product within Agent Bricks platform, the system translates natural language queries into multi-step search plans across structured and unstructured data. Tests using GPT-5, Claude-4.5, and a custom 4B-parameter model showed 35-50% accuracy gains on enterprise benchmarks, addressing the core challenge that agents often fail due to poor data retrieval rather than reasoning limitations.

Snowflake and Anthropic Sign $200M Multi-Year Agent Partnership

Snowflake and Anthropic formalized a $200 million expansion extending Claude models across 12,600+ enterprise customers for deploying AI agents within corporate data environments. The partnership focuses on agents that analyze structured and unstructured data across financial, operational, and customer datasets while maintaining governance controls critical for regulated industries. Thousands of customers already process large data volumes monthly through existing Claude integrations; this agreement scales production deployments with multi-agent collaboration capabilities.

Exabeam Launches Industry-First AI Agent Behavior Analytics

Exabeam introduced the first connected system for AI agent security, extending UEBA to monitor autonomous agent activity alongside traditional user behavior. The January New-Scale release delivers Agent Behavior Analytics with behavioral detections for suspicious deviations, unified timeline-driven investigation tools, and AI Agent Security Posture visibility. Nearly 80% of enterprises now deploy AI agents, creating attack surfaces where malicious prompts can trigger insider threats at machine speed—Exabeam predicts AI agent oversight will become a core security category in 2026 alongside identity and cloud protection.

Visa and Akamai Partner on Trusted Agent Protocol for Agentic Commerce

Visa and Akamai integrated Visa's Trusted Agent Protocol with Akamai's edge-based behavioral intelligence to authenticate AI shopping agents and prevent fraud across 175 million merchant locations. The collaboration addresses the dual-identity challenge of verifying both the agent and the consumer it represents, enabling merchants to differentiate legitimate automated traffic from malicious bots. With traffic from AI sources surging 1,200% for retailers, the protocol provides standardized agent authentication, intent signaling (browsing vs. paying), and secure payment interactions through existing checkout flows.

Yahoo DSP Launches Agentic AI Capabilities at CES 2026

Yahoo DSP announced integration of Agentic AI at CES 2026, transitioning platforms from predictive to proactive optimization with agents automating testing, pacing, and cross-channel orchestration. DSPs now function as orchestration engines with transparent, privacy-safe controls, with additional advanced agents rolling out throughout 2026. The announcement signals industry shift toward autonomous campaign management across digital advertising channels.

What to Watch this Week

 🎙️AI on Fire: Real Builders. Real Heat.

Stories from founders building AI Agents. Explore all episodes here

⚔️ AArena: The Battleground for AI

Stop demo-hopping. One workspace. Every agent. Real results.

👉 Enter the Arena | 👉 See the Battleboard

🏆 The Leaderboard Never Sleeps

The global ranking of AI agents is shifting every day. Who’s on top? Who just dropped?

👉 Explore the Leaderboard →

🗺️ The Map of AI Agents (Live & Growing)

We’re charting the entire AI agent ecosystem — thousands of options across categories.
Your next agent is already on the map.

👉 Explore the Landscape Map →