AI Wiki · Mediocre Security
Company

Google

Alphabet's core company; parent of Google DeepMind, maker of Gemini and TPUs, investor in Anthropic and creator of the A2A protocol.

Google (Alphabet) is the parent of Google DeepMind and sells Gemini across Search, Workspace and Google Cloud. It designs TPUs (now reachable from PyTorch via TorchTPU; on Nov 9 2026 it added the TPU 8tx, with hardware confidential computing for private fine-tuning and agent execution, and the dual-chip TPU 8i Max, making four eighth-generation TPU systems; it says nearly 100 customers use TPUs), created the A2A protocol and the Universal Commerce Protocol, ships the Agent Development Kit, and is a major investor in and cloud supplier to Anthropic. Google agreed in 2025 to acquire Wiz. It paid XBOW a $250K bounty for an AI-built Chrome exploit chain and patched 62 CVEs from Anthropic’s red-team chains. Legal: a CJEU Advocate General said Gemini can infringe publishers’ rights; a US judge dismissed antitrust suits over AI Overviews; North Carolina parents sued in November 2026 alleging Gemini on a school Chromebook worsened their teen’s mental-health crisis (allegations). Infrastructure: a $17B first Wyoming data center powered partly by 702 MW of fuel cells with carbon capture (Nov 5 2026), on a site it bought from Crusoe (confirmed Nov 6). Its venture arm is GV.

Its Threat Intelligence Group (GTIG) publishes a quarterly AI threat tracker; the Nov 2026 edition reported the first fully autonomous exploit pipeline it had seen in the wild, exploit_agent.

Timeline

Every dated event in the wiki that involves Google, newest first.

  • Policy
    Google now keeps Gemini chats about two months even with history turned off

    Google's updated Gemini Apps Privacy Hub raises retention for chats made with Gemini History off from 72 hours to around two months, used to respond, process feedback and for safety, with help from human reviewers. Gizmodo compares OpenAI (up to 30 days for temporary chats) and Anthropic (30 days by default). Google declined to comment beyond a blog post on updated Gemini controls.

  • Research
    Google researcher's AI-agent 'discovery' of a Maya site turns out to be already published

    A Google researcher said a team of AI agents working on lidar data had found a possibly unmapped Maya site in Mexico; archaeologists pointed out it had appeared in an academic journal the year before. The Verge uses the episode to describe scientists' frustration with amateurs making AI 'discoveries' outside their field.

  • Release

    A cheaper Gemini 4 tier at $1.50 / $7.50 per million tokens with a 1M-token context. Google cites 51 on the Artificial Analysis Intelligence Index (Argon: 50) and 2.3x more resilience on Gray Swan's indirect Prompt injection benchmark (vendor claims).

    Sources: Google, [link]
  • Release
    Google launches the Gemini Decisions API

    Gemini Decision 1 scores options in a single pass instead of generating text: $0.10 per million input tokens, no output charge, 150-250 ms latency, up to 1,000 questions per request. Google claims the lowest calibration error on its benchmarks (vendor claim).

  • Release
    Google makes Gemini 4 Carbon generally available

    Carbon, first released to cyber defenders in the Fairwind Program two weeks earlier, opens to paying developers and enterprises via the API and rolls out to AI Ultra subscribers. Introductory price $2 / $10 per million tokens until Nov 30, then $4 / $20. Google claims top ranks on CursorBench 4.0 and several agent and cyber benchmarks (vendor claims).

  • Fundingreported, not confirmed~$5B at ~$50B valuation (reported; up from a $2.1B round five months ago)
    Isomorphic Labs: Alphabet's DeepMind spinout is in talks to raise about $5B

    REPORTED talks only: Alphabet's DeepMind spinout is in talks to raise about $5B; terms could change.

    Sources: Bloomberg (via AI Weekly)
  • IPOOct 14-21 2026Valuation talk ~$1.5T, down from ~$2T (reported)
    Anthropic: Pre-IPO (reported, not filed)

    Pre-IPO (reported, not filed): Q3 revenue ~$16.8B (+45% q/q) and adjusted-profitable per attendees of an Oct 14 investor meeting; WSJ says the draft prospectus shows $16.3B H1 revenue and $600B+ in infrastructure commitments; FT says it offered Blackstone a 10%+ yield to anchor $9B of a $60B Broadcom-arranged debt package for its Google TPU lease; IPO expected around mid-November; may commit ~$45B more to Nscale

    Sources: Bloomberg; CNBC; Reuters; FT; WSJ (via The Neuron)

Connections

Google and everything one link away. Drag to move, scroll to zoom, click a node to open it.Open in the full graph →

What links here

Protocol Agent2Agent (A2A) · Google Universal Cart (Universal Commerce Protocol) · Model Context Protocol (MCP) · x402

Concept AI and CVEs: latent flaws, now exploitable

Legal action AI copyright litigation (2026) · Nonprofits v. frontier AI labs

AI lab Anthropic · Google DeepMind

Incident Anthropic red team's AI-built exploit chains · exploit_agent framework · Gemini CTF incident (Irregular) · OpenCode Console API-key theft · PaperCut AI agent campaign

Company Apple · Crusoe · Fireworks AI · Isomorphic Labs · Okta · Shopify · Tenspresso · Waymo · Wiz · XBOW

Person Demis Hassabis · Sundar Pichai

Model family Gemini · Gemma

Product Google Agent Development Kit (ADK) · Google TorchTPU · IBM watsonx Orchestrate · OpenCode

Investor GV (Google Ventures)

Organization Irregular

Standard SAFA (Standards Authority for Frontier AI) · White House Accord on Super Intelligence

Government body US Department of Defense · US State Department

Cited by

Cited as a source by 11 timeline events.

2 articles cite Google as a source: Gemini · Google Agent Development Kit (ADK)

Sources

  1. Wikipedia, “Google”
  2. Crusoe, “Crusoe Announces Wyoming Data Center Campus Update”, Nov 6 2026
  3. The Next Web, “Google unveils TPU 8tx and 8i Max chips for secure and long-context AI”, Nov 9 2026
  4. SiliconANGLE, “Google says an AI framework turned security patches into exploits with no human in the loop”, Nov 16 2026