AI News, 11 June 2026

Today’s most substantial item is OpenAI’s account of IT firm Endava rebuilding its software delivery around AI agents, though without concrete detail in the available material. The rest of today’s roundup is a set of bare version bumps from Claude Code, Gemini CLI, OpenAI Codex and OpenClaw, plus a Google piece on using AI search for second-hand and vintage shopping, all backed by very thin source material.
Today’s Stories
Section titled “Today’s Stories”1. OpenAI: How Endava is redesigning software delivery around AI agents
Section titled “1. OpenAI: How Endava is redesigning software delivery around AI agents”What changed: OpenAI published the update on 3 Jun 2026. OpenAI has published a customer story describing IT services firm Endava rebuilding its software delivery around AI agents. Beyond the headline, the available material offers no description of what the change actually involves.
Why it matters: The signal here is that agent-based development is now being practiced inside large IT organizations, which may interest people experimenting with agents at home, but without figures on time saved or changed workflows there is nothing concrete to carry over to your own setup.
How to verify it: Look up the full customer story on OpenAI’s site and check whether it contains concrete figures or method descriptions not captured in this summary.
Uncertainty: Which AI agents or tools Endava actually uses, how far-reaching the change is, and whether the results have been checked by anyone other than OpenAI itself are all absent from the source material.
Source: How Endava is redesigning software delivery around AI agents
2. Google AI: 5 ways Google Search can level up your thrift and vintage shopping
Section titled “2. Google AI: 5 ways Google Search can level up your thrift and vintage shopping”What changed: Google AI published the update on 2 Jun 2026. A Google article outlines five ways to use AI features in Google Search and Google Shopping to find second-hand and vintage items. The five specific tips are not included in the source material available here.
Short source excerpt: Uncover second-hand scores with AI tools in Google Search and Shopping. Try these 5 thrifting tips from Google Skip to…
Why it matters: If the features are already live, they could make it easier to track down used furniture or electronics for a smart-home project, but without the concrete steps it is impossible to judge how much effort or which app version is required.
How to verify it: Try searching for a used or vintage product yourself in Google Search or the Shopping app, and see whether filters or AI-generated suggestions appear that you have not seen before.
Uncertainty: Whether these are entirely new features or already-known ones, which countries they cover, and whether an updated app is required cannot be determined from the available material.
Source: 5 ways Google Search can level up your thrift and vintage shopping
3. OpenClaw: openclaw 2026.6.5
Section titled “3. OpenClaw: openclaw 2026.6.5”What changed: OpenClaw published the update on 8 Jun 2026. OpenClaw, an open-source personal AI assistant project, has shipped version 2026.6.5. The source material consists solely of the version number, with no list of changes.
Short source excerpt: Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞 - Release openclaw 2026.6.5 · openclaw/openclaw Release…
Why it matters: People running OpenClaw locally cannot tell from this summary whether the update brings new features, bug fixes, or security changes, so they should check the changelog themselves before updating a running system.
How to verify it: Open the openclaw/openclaw release page on GitHub and read the full changelog for 2026.6.5.
Uncertainty: What concretely changed, whether the update is backward compatible, and whether it introduces new dependencies all fall outside what the source shows.
Source: openclaw 2026.6.5
4. Claude Code: v2.1.172
Section titled “4. Claude Code: v2.1.172”What changed: Claude Code published the update on 9 Jun 2026. Claude Code, Anthropic’s terminal-based coding tool, has been updated to version 2.1.172. The material shows only the version number, with no list of changes.
Short source excerpt: Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster…
Why it matters: For daily Claude Code users, a new version usually means adjustments to commands or behavior, but without a changelog there is no way to see whether something you depend on has changed.
How to verify it: Run the version command in your terminal and compare it against the release notes for anthropics/claude-code on GitHub to see what the update actually contains.
Uncertainty: Whether the update fixes security issues, adds features, or merely tweaks details cannot be determined when the changelog text is missing.
Source: v2.1.172
5. Gemini CLI: Release v0.47.0-preview.0
Section titled “5. Gemini CLI: Release v0.47.0-preview.0”What changed: Gemini CLI published the update on 9 Jun 2026. Google has put out a preview build of Gemini CLI, its open-source terminal tool for Gemini, labeled 0.47.0-preview.0. No changelog accompanies it in the source material.
Short source excerpt: An open-source AI agent that brings the power of Gemini directly into your terminal. - Release Release v0.47.0-preview.0 · google-gemini/gemini-cli…
Why it matters: The preview label means users should expect instability or in-progress changes rather than a finished feature, and without a changelog it is hard to judge whether it is safe to use in a stable setup.
How to verify it: Check the release page for google-gemini/gemini-cli on GitHub to read the full changelog and see how close the features are to a stable release.
Uncertainty: What this preview actually adds or changes, and when the content is expected to land in a stable release, cannot be read from the source.
Source: Release v0.47.0-preview.0
6. OpenAI Codex: 0.139.0
Section titled “6. OpenAI Codex: 0.139.0”What changed: OpenAI Codex published the update on 8 Jun 2026. OpenAI has shipped version 0.139.0 of Codex, its terminal-based coding tool. The source text does not extend beyond the version number.
Short source excerpt: Lightweight coding agent that runs in your terminal - Release 0.139.0 · openai/codex Release 0.139.0 · openai/codex · GitHub /…
Why it matters: With Codex built into a workflow, a new version can change commands or default behavior, but this summary cannot point to anything concrete that was fixed or added.
How to verify it: Read the openai/codex 0.139.0 release notes on GitHub before updating, to check whether the changes affect existing scripts or configuration.
Uncertainty: The content of the update, whether it breaks compatibility with earlier versions, and whether known bugs were fixed all lie outside what the source documents.
Source: 0.139.0
Why It Matters
Section titled “Why It Matters”- The focus is changes that can affect AI products, models, browser experiences, agents, API usage, privacy, pricing, or security.
- A new feature is useful in the home only after access, logging, and rollback have been tested.
- The article separates documented information from editorial judgment and open questions.
Sources and Further Reading
Section titled “Sources and Further Reading”| Publisher | Source | Date |
|---|---|---|
| OpenAI | How Endava is redesigning software delivery around AI agents | 3 Jun 2026 |
| Google AI | 5 ways Google Search can level up your thrift and vintage shopping | 2 Jun 2026 |
| OpenClaw | openclaw 2026.6.5 | 8 Jun 2026 |
| Claude Code | v2.1.172 | 9 Jun 2026 |
| Gemini CLI | Release v0.47.0-preview.0 | 9 Jun 2026 |
| OpenAI Codex | 0.139.0 | 8 Jun 2026 |
Editorial Note
Section titled “Editorial Note”SmartBolig.net used automated research, deduplication, and drafting for this issue. Editorial rules require a primary source, verifiable links, and explicit uncertainty.
Comments
SmartBolig // AI Core
SmartBolig AI
Get help with troubleshooting, architecture and concrete configuration. I use broad AI knowledge and find SmartBolig guides when they improve the answer.
Start with a useful question