Building a Context Pruning Pipeline for Long-Running Agents

As AI agents move from experiments to production, AWS, Cloudflare, and others are redesigning cloud infrastructure for a future dominated by machine-generated internet traffic instead of human users.

Researchers in China built a model that uses an AI’s downtime to prepare for users’ next question before they ask it.

Google DeepMind CEO Demis Hassabis believes progress toward artificial general intelligence (AGI) is moving faster than expected and that society now has only a few years to prepare. He believes AGI could arrive around 2030, though acknowledges it could be here in 2029 — or even sooner. In an interview with Axios, Hassabis said that today’s AI agents — systems capable of performing tasks independently — should be viewed as a sort of “practice run” for significantly more powerful AI in the future. He also warned that governments, economists, and society at large are not taking this development seriously enough. One particular risk he highlighted is that AI systems in the future might begin to improve their own development. “All the leading labs are pretty focused on that,” Hassabis told Axios. “It will yield clear benefits in the form of faster research. But there are also risks associated with that type of system.”

Google Pay is overhauling its payment infrastructure for an impending wave of transactions from AI agents. The latest updates introduce the Universal Commerce Protocol and a new server architecture, positioning Google Pay as a central clearinghouse for purchases executed by autonomous agents rather than human users. AI agents – designed to perform tasks like booking […] The post Google Pay preps for AI agents with Universal Commerce Protocol appeared first on AI News.

Snowflake said it plans to acquire US-based startup Natoma to boost governance, security, and connectivity for AI agents operating across heterogeneous enterprise environments, amid growing efforts by organizations to move agentic AI workflows from pilots into production. The cloud data platform provider is betting that enterprises will increasingly require centralized governance, identity controls, and auditability as AI agents begin interacting more deeply with internal applications, APIs, and business workflows through the emerging Model Context Protocol (MCP) standard, an area in which Natoma claims to specialize. Natoma’s platform, which provides MCP-based tool access along with governance and observability capabilities, will be integrated into Snowflake to help enterprises securely connect Cortex Agents, Snowflake Intelligence, Cortex Code, and other AI platforms with enterprise systems spanning SaaS applications, cloud environments, VPCs, and on-premises infrastructure through M

Attackers drained an estimated $200,000 from DeFi liquidity pools on Ethereum — specifically Uniswap V3 — after exploiting weaknesses in the WUSD.fi and GLOVE incentive system, according to security researchers at ExVul. The attackers cycled funds through multiple wallets to repeatedly farm rewards, taking advantage of flaws baked into the protocol’s incentive structure. Related Reading: […]

The rapid uptake of agentic AI has exposed a range of issues with our non-deterministic helpers. That’s mainly because AI agents are not people and don’t behave like people, even though they generally use the same APIs as humans. For one thing, they make many more queries than a human would, as they build the necessary context to deliver a response. Anecdotal data from companies that have worked with agents or who have users who access services through agents indicate that this can mean massive increases in API usage, which have affected availability. This increase is the result of automated requests flooding in and blocking calls and responses from APIs that worked perfectly well a year or so ago but now are struggling to cope with the load. A fundamental redesign of our APIs is necessary, but budgets, resourcing, and capacity make this hard to deliver overnight. What’s needed, then, is a way to manage agent interactions with APIs, treating agents as a new class of user, providing and

Robinhood is enabling AI agents to trade stocks and make payments on users’ behalf, marking a significant push into agentic finance. Users can now create dedicated accounts for their AI agents, preload them with funds, and connect them via the platform’s Model Context Protocol server to execute trades, analyse portfolio risk, and identify new investment […]