Urgent.News

What's breaking now, across thousands of outlets.

AI

Glean unveils Tau desktop workspace, claims token-cost edge over Claude

Glean Technologies Inc. today unveiled Glean Tau, a desktop workspace that connects the company’s enterprise artificial intelligence to a user’s local files, applications and code. The launch anchors a broad slate of product news at Glean:GO, the company’s conference this week in San Francisco. Packaged with it were benchmark numbers aimed at Anthropic PBC. Glean said […] The post Glean unveils…

Glean unveils Tau desktop workspace, claims token-cost edge over Claude

Glean Technologies Inc. unveiled Glean Tau, a desktop workspace, at Glean:GO conference in San Francisco. The desktop workspace connects Glean's enterprise AI to a user's local files, applications and code. Glean touted a 5.2-times token-cost advantage over Anthropic's Claude Cowork per query in testing. Users found Glean to be preferred 3.6 times more often.

Tau runs on an open-source harness and leverages Glean's enterprise context with whatever sits on the machine. The pitch emphasizes reduced time spent on file hunting, background feeding, and model steering through each step, known as "botsitting." Tau can plan multistep work, carry it out on a user's behalf, and self-check its output. Glean listed file organization, document analysis, spreadsheet building, coding, and engineering workflows among the jobs Tau handles.

Glean's cost-saving strategy involves keeping permission-aware company context available to a model, thereby avoiding repeated retrieval calls and subsequent token spending. Glean Intelligence includes centralized usage visibility, administrative controls over AI spending, and enhanced auto routing. Administrators can push work toward efficient, balanced, or frontier models based on the desired price-performance tradeoff.

Other products and expansions announced at the conference include Glean Transform, which reads activity signals to determine how work is done, recommends AI tasks, and measures business impact; an AI task management feature that drafts follow-ups and prepares users for meetings; expanded interactive dashboards that blend structured data with various data types; and improvements in collaboration, governance, and security.

Team chat brings multiple people into a shared thread or artifact, while AI Gateway is expanded to support more entry points and enforce restricted-topic policies. Context-aware threat detection draws on the Enterprise Graph to identify potential data harvesting, exfiltration, or destructive action.

Glean's products are available in different phases: Glean Intelligence, AI usage controls, and memory via MCP are generally available; beta includes independent agents, AI Gateway, auto routing, team chat, and dashboard refresh. Tau is still in development, as are Glean Transform, threat detection, task management, email triage, a meeting coach, and skills via MCP.

Six customers are mentioned in the announcement: Rhonda Baldwin, chief information officer at LaunchDarkly, and David Lee, director of people and places enablement at LegalZoom.com Inc. Both customers praised the platform for enabling faster AI scaling without compromising governance. The company is venture capital-backed, having raised $150 million in a Series F round at a valuation of $7.2 billion in June 2025.

Written by urgent.news from SiliconANGLE's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at siliconangle.com →

More in AI

Why Your AI Agent Fails at Observability: A Debugging Framework for Memory, Tool Calls, and RAG

Originally published on tamiz.pro . You built the agent. The prompt looks solid. The RAG pipeline is technically "working" because your vector DB returns results.

  • Agent failures often stem from opaque black box nature, not the LLM itself
  • Need semantic tracing to understand causal chain of tool calls and memory ops
  • Three pillars of observability: Memory State, Tool Execution, Retrieval Validity

More from Wednesday 26 August →