Hello, AI Explorer!
Good Saturday. This week: Anthropic raised $65B at a $965B valuation and launched Opus 4.8 the same day, Apple is rebuilding Siri on Google Gemini with a standalone ChatGPT competitor coming at WWDC June 8, CNN sued Perplexity for copying 17,000 stories word for word.
Welcome back to your weekly AI digest: human-curated, zero fluff.
Every Wednesday and Saturday, we bring you:
- The week's most important AI developments (no fluff)
- Hand-picked AI tools actually worth your time
- Expert insights on what these changes mean for you
Let's dive into this week's discoveries! ⚡
🔥 This Week in AI
🔥 This Week in AI |
May 30, 2026 |
| 💰 | Anthropic hits $965B valuation -- surpasses OpenAI for the first time after $65B Series H |
| 🧠 | Claude Opus 4.8 launches same day -- sharper reasoning, honesty improvements, new dynamic workflow tool |
| 🍎 | iOS 27 Siri is being rebuilt on Google Gemini -- standalone ChatGPT-style app revealed ahead of WWDC |
| ⚖️ | CNN sues Perplexity for copying 17,000 stories verbatim -- first TV network AI copyright lawsuit |
| 🎬 | YouTube auto-labels photorealistic AI videos now -- even when creators don't disclose it |
| 🏛️ | AI cracked a 400-year-old Vatican cipher in 29 minutes -- 408 pages of strange medical secrets unlocked |
| 📊 | Cursor report: top 1% of AI-assisted developers now produce 46x more code per week than the median dev |
| 🔍 | Google I/O: Gemini 3.5 Flash, Gemini Omni, and the biggest Search overhaul in 25 years |
| ⚡ | 4 AI Tools This Week: Bluedot, Emergent, Wispr Flow, Rows -- all inside. |
ANTHROPIC
| 💰 |
Anthropic Hits $965B Valuation -- and Launches Opus 4.8 on the Same Day |
On May 28, Anthropic closed a $65 billion Series H round at a $965 billion post-money valuation -- the first time an AI startup has eclipsed OpenAI in market value. The round was co-led by Altimeter Capital, Sequoia Capital, Dragoneer, and Greenoaks. Samsung, SK Hynix, and Micron joined as infrastructure partners. OpenAI last raised at an $852 billion valuation in March 2026.
Anthropic also shipped Claude Opus 4.8 the same day -- 41 days after Opus 4.7. The update improves agentic coding and multidisciplinary reasoning. Early testers note the model is more likely to flag its own uncertainties and less likely to assert things it cannot verify. A new "dynamic workflow" tool handles complex multi-step automations more reliably than previous versions.
What happened:
• $65B Series H co-led by Altimeter, Sequoia, Dragoneer, Greenoaks -- $965B post-money valuation
• Surpasses OpenAI's $852B valuation from March 2026 for the first time
• Opus 4.8 ships same day: stronger reasoning, honesty improvements, dynamic workflow tool
Why it matters:
Anthropic sitting above OpenAI in market cap would have seemed unlikely 18 months ago. The release cadence -- Opus 4.8 shipping 41 days after 4.7 -- tells you the pace of model development is no longer measured in quarters. For anyone building on Claude's API, the dynamic workflow tool is the piece worth testing: it means more reliable multi-step automations without the prompt engineering overhead.
APPLE -- WWDC 2026
| 🍎 |
iOS 27 Siri Is Getting a Google Gemini Backbone and a Standalone ChatGPT Competitor |
Leaked screenshots and pre-WWDC reports reveal Apple is rebuilding Siri from the ground up for iOS 27. The new Siri runs on a custom model built on Google Gemini technology -- Apple is reportedly paying around $1 billion per year for the license. A standalone Siri app will feature chat history, file and photo uploads, and a chatbot-style interface accessible from the Dynamic Island or a swipe-down gesture.
Apple is also introducing an "Extensions" system in iOS 27 that lets users set any third-party AI -- Claude, Gemini, ChatGPT -- as the default engine for Siri, Writing Tools, and Image Playground system-wide. Developer betas are expected at WWDC on June 8.
What's coming in iOS 27:
• Siri rebuilt on Google Gemini -- Apple paying ~$1B/year for the license
• Standalone Siri app with chat history, file uploads, Dynamic Island access
• Extensions: set Claude, ChatGPT, or Gemini as your iOS AI default system-wide
Why it matters:
If Apple ships this, Gemini becomes the backbone of the most-used mobile assistant in the world. The Extensions system is arguably bigger news -- it turns iOS into an open AI platform. Power users could set Claude as their default writing AI across every Apple app on day one. Watch WWDC June 8 for whether this stays as leaked or gets cut.
| ⚖️ |
CNN Sues Perplexity for Copying 17,000 Stories Verbatim |
CNN filed a federal copyright lawsuit against Perplexity on May 28, alleging the AI search engine copied more than 17,000 CNN stories, videos, and images to power its products. This is the first AI copyright lawsuit from any television network. Perplexity's CCO responded: "You can't copyright facts." CNN says it tried to negotiate a licensing deal last year before blocking Perplexity's scraper entirely.
WHY IT MATTERS:
CNN joins the NYT, Reddit, and Dow Jones in suing Perplexity. The "can't copyright facts" defense has a history, but verbatim article copying is a different argument. The ruling will determine how AI search tools can legally use news content going forward.
| 🎬 |
YouTube Now Auto-Labels AI-Generated Videos -- Whether Creators Disclose It or Not |
YouTube is rolling out automatic detection of "significant photorealistic AI use" across all uploads starting May 2026. Long-form videos get the label directly below the player. Shorts get an overlay on the video itself. Content that is clearly stylized or fantastical gets a quieter label in the description. For videos made with YouTube's own tools (Veo, Dream Screen) or with C2PA metadata, the AI label is permanent.
WHY IT MATTERS:
This closes the disclosure gap creators were exploiting. A platform can now algorithmically override a creator's choice not to label. Expect TikTok and Instagram to follow. If you produce AI-assisted video, automatic labeling is now the default assumption on major platforms.
|
⚡ Coming in Hot AI Tools of the Week |
May 30, 2026 |
4 tools worth adding to your workflow right now.
| 🎙️ |
AI MEETING NOTES Bluedot Bluedot records your meetings through a Chrome extension -- no bot joins the call, no visible participant icon. It works across Google Meet, Zoom, Teams, and Webex on any operating system, identifies speakers automatically, and syncs notes to Salesforce or HubSpot on paid tiers. 5 free meetings per month, then from $18/month. Try Bluedot → |
| ⚡ |
VIBE CODING Emergent YC-backed full-stack app builder that turns a plain-language description into a working app -- frontend, backend, database. 1.5M+ users have shipped real products without touching a code editor. Build with Emergent → |
| 🗣️ |
AI DICTATION Wispr Flow Voice dictation that works in any app you can type in -- emails, Slack, Google Docs, your IDE. Wispr cleans up filler words and grammar automatically as you speak, so the output reads like written text, not a transcript. Most users dictate 3x faster than they type. Free tier available, supports 100+ languages. Try Wispr Flow free → |
| 📊 |
AI PRODUCTIVITY Rows A spreadsheet where you ask data questions in plain English instead of writing formulas. Type "show me our top 10 customers by revenue last month" and it writes and runs the query. Connects directly to Stripe, HubSpot, or a database. The AI data analyst your team cannot yet afford to hire. Freemium. Discover Rows → |
⚡ Quick Hits
📊 Cursor report: top 1% of devs produce 46x more code per week than the average
Cursor's Developer Habits Report shows the productivity gap is growing fast. The top 1% of users produce 46x more lines per week than the median active developer and merge 15x more PRs. Since early 2026, AI-written code that survives in the codebase past the 60-minute mark rose from 76% to 81%. See the full report →
🔍 Google I/O: Gemini 3.5 Flash, Gemini Omni, and the biggest Search box upgrade in 25 years
Google shipped Gemini 3.5 Flash (frontier-class intelligence at Flash speeds) and Gemini Omni (any input to any output, starting with full video). CEO Sundar Pichai declared the start of the "agentic Gemini era" -- AI that takes actions, not just answers questions. Google Search also added text, image, file, video, and Chrome tab inputs. See all 100 announcements →
🏢 KPMG deploys Claude to 276,000 employees across 138 countries
KPMG rolled out Claude across its global audit, tax, advisory, and consulting workflows -- one of the largest enterprise AI deployments ever reported by a professional services firm. The firm covers 138 countries and 276,000 staff. The deployment is focused on document analysis, research, and drafting. Read more →
📋 OpenAI published its Frontier Governance Framework on May 29
OpenAI outlined its policy approach for managing risks as models approach AGI-level capabilities. The framework covers evaluation thresholds, deployment gates, and research transparency. No new model releases were tied to the announcement -- this is a policy document, not a product launch. Read the framework →
That's all for this edition! What tools are you most excited to try?
Reply to this email - we read every response and use your feedback to improve future editions.
See you on Wednesday,
The AI Tool Discovery Team
Feedback
What'd you think of today's newsletter?
Vote below to let us know how we're doing.