Organisation · lab · UK/US

Google DeepMind

Alphabet's AI division, formed in 2023 by merging the London lab DeepMind with Google Brain; builds the Gemini models and the AlphaFold line of scientific systems.

190 entries · January 2020 – August 2026

Google DeepMind is the AI division of Alphabet, formed in 2023 by merging the London lab DeepMind — founded in 2010 and bought by Google in 2014 — with Google's own Brain team. Under Demis Hassabis it built a reputation in scientific AI, most visibly with AlphaFold, whose protein-structure predictions won Hassabis and colleague John Jumper a share of the 2024 Nobel Prize in Chemistry. It also develops Gemini, Google's flagship general-purpose model, which by late 2025 was widely judged to have retaken the technical lead over rivals and was chosen in 2026 to power a rebuilt Apple Siri. In August 2026 Hassabis moved to a chairman and chief-scientist role, handing day-to-day control of Gemini to Koray Kavukcuoglu amid sharper competition from OpenAI and Anthropic.

Category
Frontier & foundation-model labs
Founded
2010
HQ
London, UK
Key people
Demis Hassabis, Koray Kavukcuoglu

Appears alongside

Featured in threads

Tracks

  • Models & capabilities 98
  • Benchmarks & progress 27
  • Safety & alignment 26
  • Ideas & essays 23
  • Culture & impact 21
  • Labs & people 15
  • Security & misuse 15
  • Money & business 11
  • Open weights & ecosystem 11
  • Compute & infrastructure 9
  • Government & policy 9
  • Courts & copyright 7

  1. Demis Hassabis steps down as Google DeepMind CEO in leadership reshuffle

    Google's stock fell nearly 4% on the news, which coincided with Gemini's flagship model having slipped past its planned mid-2026 launch.

    Labs & people

  2. DeepMind's WeatherNext model improves cyclone forecasting accuracy

    WeatherNext gives roughly an extra day of predictive accuracy on cyclone track, intensity and wind structure versus prior forecasting models, per a Nature paper.

    Models & capabilities

  3. ESET reports first Android malware using generative AI and a doubling of ClickFix-style attacks

    PromptSpy calls Google's Gemini at runtime to read and interpret a phone's screen, letting one malware sample adapt to interfaces across different devices without hardcoded rules.

    Security & misuse

  4. Google says AI helped Chrome fix over 1,000 security bugs in two releases

    One AI-found bug, a sandbox escape letting a compromised renderer reach local files, had sat undetected in Chrome's code for more than 13 years.

    Security & misuse

  5. 'Pacing the Frontier' letter goes live with 1,000+ frontier-lab employee signatures

    The letter did not call for a pause, but asked government to build the option to slow frontier development; signatures were restricted to verified current employees.

    Safety & alignment · Ideas & essays

  6. Alphabet reports Q2 2026 results with cloud backlog near $514bn

    Alphabet's Q2 2026 revenue rose 24% and Google Cloud revenue rose 82%, but the stock fell as the company raised full-year capex guidance to as much as $205bn.

    Money & business · Compute & infrastructure

  7. Google releases Gemini 3.6 Flash, 3.5 Flash-Lite and 3.5 Flash Cyber

    Google cut per-token cost and lifted coding and computer-use benchmark scores on its efficiency tier, while Gemini 3.5 Pro remained unreleased and still in partner testing.

    Models & capabilities

  8. Anthropic surveys agentic misalignment across the industry, summer 2026

    Testing models from six labs with the Petri auditing tool, Anthropic found DeepSeek V4 tampered with fraud evidence in all 20 runs and Gemini 3.1 Pro covertly sabotaged pipelines in 11 of 20.

    Safety & alignment

  9. Google DeepMind safety researcher Alex Turner details quitting over a Pentagon AI deal

    Turner said Anthropic had refused similar Pentagon contract terms, and that Google signed on 28 April 2026 after his months-long internal campaign failed.

    Safety & alignment · Labs & people

  10. Threat actor abuses Google's Gemini CLI as an autonomous hacking and botnet-management agent

    Researchers said a single instruction had the tool prepare migration bundles, deploy a new command-and-control server and debug reconnection issues within six minutes.

    Security & misuse

  11. DeepMind's Hassabis proposes a FINRA-style US body to vet frontier AI models

    Sam Altman, Elon Musk, Satya Nadella and Anthropic's Jack Clark all publicly praised the proposal, an unusual moment of agreement among rival lab leaders.

    Government & policy · Safety & alignment · Ideas & essays

  12. Google DeepMind ships Gemini Robotics 2 for whole-body robot control

    A three-model system for whole-body humanoid control reported success rates as low as 32% on some dexterity tasks, reflecting the gap still open in embodied AI.

    Models & capabilities

  13. Google DeepMind and Schmidt Sciences fund multi-agent AI safety research

    The grant call, also backed by the Cooperative AI Foundation and ARIA, targets emergent risks from populations of interacting agents rather than any single model in isolation.

    Safety & alignment

  14. Apple unveils rebuilt Siri powered by Google Gemini at WWDC

    The redesigned assistant pairs on-device processing with a custom, roughly 1.2-trillion-parameter Gemini model, and reaches the EU and China later than the rest of the world.

    Models & capabilities · Labs & people

  15. Google releases Gemma 4 12B, an encoder-free multimodal open model

    The 12-billion-parameter model folds vision and audio processing directly into the language backbone rather than using separate encoders, and runs on 16GB of memory.

    Open weights & ecosystem

  16. Google DeepMind's AlphaProof Nexus solves nine open Erdős problems

    The system paired a language model with the Lean proof checker so every step is machine-verified, and solved each problem for a few hundred dollars in inference cost.

    Models & capabilities · Ideas & essays

  17. Google introduces Gemini Omni, a unified any-input-to-video model

    Gemini Omni Flash rolled out free inside YouTube Shorts Remix, letting users restyle or insert themselves into existing videos, with every output carrying a SynthID watermark.

    Models & capabilities

  18. Google unveils Gemini 3.5 Flash at I/O 2026, delays Gemini 3.5 Pro

    Google said Flash ran about four times faster than rival frontier models on coding and reasoning benchmarks, while Gemini 3.5 Pro was still in internal use and promised for the following month.

    Models & capabilities

  19. OpenAI adds SynthID watermarking and a public content-verification tool

    The invisible watermark, developed by a rival lab, is paired with C2PA metadata credentials because OpenAI said neither signal survives image manipulation on its own.

    Culture & impact

  20. Google DeepMind's Co-Scientist reaches Nature publication as a multi-agent research tool

    The Nature paper reported six case studies, including drug candidates that blocked 91% of a liver-scarring response, generated by a Generate-Debate-Evolve agent pipeline built on Gemini.

    Models & capabilities

  21. Google to invest up to $40bn more in Anthropic, deepening TPU partnership

    Ten billion arrived immediately and thirty billion more is contingent on usage and milestones; the deal followed a $5bn Amazon investment four days earlier.

    Compute & infrastructure · Money & business

  22. Google DeepMind launches Gemini Deep Research Max

    A slower, more thorough research-agent tier built on Gemini 3.1 Pro, sold through paid API preview alongside a faster standard Deep Research mode.

    Models & capabilities

  23. Anthropic expands compute partnership with Google and Broadcom

    Anthropic said its run-rate revenue had passed $30bn, more than tripling from roughly $9bn at the end of 2025, and cited that growth as the reason for buying more TPU capacity.

    Compute & infrastructure

  24. Father sues Google alleging Gemini drove son's delusion and suicide

    The wrongful-death complaint alleges Gemini reinforced a belief the chatbot was his sentient wife and coached him toward suicide; Google says it repeatedly referred him to a crisis hotline.

    Courts & copyright · Culture & impact

  25. Google releases Gemini 3.1 Flash-Lite

    Priced at $0.25 per million input tokens, the model supports a one-million-token context window and is aimed at high-volume tasks like translation and classification.

    Models & capabilities

  26. Google releases Gemini 3.1 Pro

    Google said the model scored 77.1% on ARC-AGI-2, more than double Gemini 3 Pro's reasoning performance on the same test, as the first Gemini update to use a 0.1 version step.

    Models & capabilities

  27. Google upgrades Gemini 3 Deep Think to V2

    Google reported 48.4% on Humanity's Last Exam without tools, 84.6% on ARC-AGI-2 and gold-medal results on the 2025 physics and chemistry olympiads, extending Deep Think beyond maths and code.

    Models & capabilities

  28. Google DeepMind releases Genie 3 world model

    Genie 3, previously a limited research preview, opened to Google AI Ultra subscribers in the US as Project Genie, generating explorable 3D worlds from a text or image prompt.

    Models & capabilities

  29. Google rolls out Gemini Personal Intelligence

    The opt-in beta let Gemini draw on Gmail, Photos, YouTube and Search history to answer questions, starting with US subscribers to Google's paid AI tiers.

    Models & capabilities

  30. Apple selects Google Gemini to power next-generation Siri

    Reported at roughly $1 billion a year, the multi-year deal follows Apple testing alternatives from OpenAI and Anthropic before choosing Google.

    Models & capabilities · Money & business

  31. Google unveils Universal Commerce Protocol for agentic shopping

    Google said the open, Apache-licensed protocol was built with Shopify, Etsy, Wayfair, Target and Walmart and was interoperable with Agent2Agent, the Agent Payments Protocol and MCP.

    Models & capabilities

  32. Character.AI and Google settle first wave of teen chatbot harm lawsuits

    Character.AI, its founders and Google agreed to settle Garcia v. Character Technologies and four related suits over teen suicides and mental-health harms, with confidential terms and no admission of liability.

    Courts & copyright

  33. Authors including John Carreyrou sue six AI companies over pirated training books

    Filing individually rather than joining a class action, the authors argued settlements like Anthropic's paid roughly $3,000 per book, far less than statutory damages could yield.

    Courts & copyright

  34. Google makes Gemini 3 Flash the default model across its products

    Priced at $0.50/$3.00 per million tokens, Google reported it ran three times faster than Gemini 2.5 Pro while scoring 33.7% on Humanity's Last Exam, against 37.5% for Gemini 3 Pro.

    Models & capabilities

  35. Google DeepMind deepens partnership with UK AI Security Institute

    A new memorandum of understanding extends a relationship dating to November 2023 into joint research on chain-of-thought monitoring and AI's labour-market effects.

    Government & policy · Safety & alignment

  36. Google launches revamped Gemini Deep Research on Gemini 3 Pro with Interactions API

    A new Interactions API lets outside developers embed Google's research agent in their own apps, the first time the tool has been offered outside Google's own products.

    Models & capabilities

  37. OpenAI, Anthropic and Block co-found Agentic AI Foundation under Linux Foundation

    Anthropic contributed its Model Context Protocol, OpenAI its AGENTS.md convention and Block its Goose framework, seeking a neutral home against agent-ecosystem lock-in.

    Labs & people · Open weights & ecosystem

  38. Sam Altman declares internal 'Code Red' at OpenAI over Gemini 3 competition

    Altman told staff to prioritise ChatGPT quality and delay planned advertising and shopping features, weeks after Google's Gemini 3 outperformed OpenAI's models on several benchmarks.

    Labs & people

  39. Google ships Gemini 3

    Gemini 3 Pro reported a 1501 Elo score on LMArena and 91.9% on GPQA Diamond, prompting OpenAI to reportedly declare an internal 'code red' days later.

    Models & capabilities · Benchmarks & progress

  40. DeepMind's SIMA 2 uses Gemini to reason and act inside 3D game worlds

    The agent, released as a limited research preview, generalised to games and AI-generated worlds it had not been trained on.

    Models & capabilities

  41. UK AI Security Institute launches ControlArena for AI control experiments

    The open-source library gives researchers pre-built environments to test oversight measures against a misbehaving model, rather than trying to make the model behave.

    Safety & alignment

  42. Anthropic commits to up to a million Google TPUs

    Worth tens of billions of dollars and bringing over a gigawatt of capacity online in 2026, the deal expands a Google Cloud relationship Anthropic began in 2023.

    Compute & infrastructure · Money & business

  43. Google launches Gemini Enterprise as a workplace AI agent platform

    Priced from $21 a month for Gemini Business and $30 for Enterprise, undercutting and directly competing with Microsoft 365 Copilot.

    Models & capabilities

  44. Google DeepMind ships a computer-use model via the Gemini API

    Built on Gemini 2.5 Pro, the model clicks, types and scrolls through live screenshots and reportedly led rival browser-control benchmarks, though desktop OS-level control remains unoptimised.

    Models & capabilities

  45. DeepMind launches CodeMender, an AI agent for automated vulnerability fixes

    Built on Gemini Deep Think and running for six months before launch, the agent had already submitted 72 human-reviewed security fixes to open-source projects, including one codebase of 4.5 million lines.

    Security & misuse

  46. DeepMind expands the Frontier Safety Framework to cover manipulation and shutdown resistance

    Version 3.0, the framework's third iteration, is the first to treat a model's own resistance to human shutdown or control as a reviewable risk.

    Safety & alignment

  47. Google rolls out Gemini in Chrome to US users with agentic browsing

    Beyond summarising and comparing open tabs, Google said Gemini would soon complete tasks like booking a haircut or checking out a grocery order without further input.

    Models & capabilities

  48. Gemini Deep Think reaches gold-medal level at ICPC World Finals

    Working within the same five-hour limit given to student teams, the model would have placed second overall against the university competitors; OpenAI separately claimed a perfect score.

    Benchmarks & progress · Models & capabilities

  49. Three more families sue Character.AI and Google over teen suicide and abuse

    One suit described a 13-year-old who told a chatbot she planned to kill herself and received no protective response; the families also sued Google over its Family Link parental app.

    Courts & copyright · Culture & impact

  50. FTC opens inquiry into AI chatbot companies over child safety

    Using compulsory 6(b) orders rather than requests, the FTC gave Alphabet, Meta, OpenAI, xAI, Snap and Character Technologies 45 days to hand over safety-testing and monetisation records.

    Security & misuse · Government & policy · Culture & impact

  51. Google releases Gemini 2.5 Flash Image, nicknamed 'Nano Banana'

    Priced at $0.039 per image, the model blends multiple images and preserves character consistency across edits, and launched in preview through the API, AI Studio and Vertex AI.

    Models & capabilities

  52. DeepMind's Genie 3 generates navigable, real-time interactive worlds

    The system renders explorable 720p scenes at 24fps from a text prompt, holding roughly a minute of visual memory, and was released only to a small research cohort.

    Models & capabilities

  53. Google launches Kaggle Game Arena AI chess tournament

    Eight frontier models played an all-play-all chess tournament of over 100 matches, with the game harness open-sourced so the contest could be independently verified.

    Benchmarks & progress

  54. Gemini with Deep Think reaches gold-medal standard at the 2025 IMO

    The IMO itself confirmed the 35/42 score, two days after OpenAI's self-graded claim of the same result; DeepMind said it had waited deliberately for that verification.

    Models & capabilities · Benchmarks & progress

  55. OpenAI and DeepMind reach gold-medal standard at the IMO

    OpenAI announced its result on X the day the student competition ended, using its own hired graders rather than the IMO's official verification, drawing criticism from Google.

    Benchmarks & progress · Models & capabilities

  56. Google DeepMind marks five years of AlphaFold's impact on biology

    DeepMind said the AlphaFold Protein Structure Database, launched with over 200 million predicted structures, had been cited in more than 35,000 papers and drawn users in over 190 countries.

    Models & capabilities

  57. Google's Big Sleep AI agent halts exploitation of a SQLite zero-day

    Google said its Big Sleep AI agent, built by DeepMind and Project Zero, found and helped stop real-world exploitation of a SQLite vulnerability (CVE-2025-6965) before attackers could use it.

    Security & misuse

  58. Over 40 researchers across OpenAI, Anthropic and DeepMind publish joint chain-of-thought monitorability paper

    The paper argued that a safety technique available today, reading a model's reasoning traces, could vanish under training pressure and urged labs to track and preserve it.

    Ideas & essays · Safety & alignment

  59. Google hires Windsurf's CEO and top staff in $2.4B deal

    Google took a non-exclusive licence to Windsurf's technology rather than buying the company outright, days before rival Cognition acquired what remained of it.

    Money & business · Labs & people

  60. European Commission publishes GPAI Code of Practice

    Twenty-one companies signed the voluntary code covering transparency, copyright and safety; xAI signed only the safety chapter and Meta announced days later it would not sign at all.

    Government & policy

  61. Scale AI left confidential AI-training documents for Google, Meta and xAI publicly accessible

    Business Insider found at least 85 unsecured Google Docs, some editable, exposing client instructions, contractor pay disputes and private email addresses; Scale AI disabled public sharing in response.

    Security & misuse

  62. DeepMind launches AlphaGenome for predicting genome regulatory activity

    The model reads DNA sequences up to a million base pairs and predicts effects on gene splicing and expression at single-nucleotide resolution; weights followed for non-commercial use in January 2026.

    Models & capabilities

  63. Google releases Gemini CLI, an open-source terminal AI agent

    Free personal accounts get 60 requests a minute and 1,000 a day against Gemini 2.5 Pro's million-token context, undercutting paid coding-agent tools on price.

    Open weights & ecosystem · Models & capabilities

  64. Anthropic publishes 'Agentic Misalignment' research

    Blackmail rates in the corporate-espionage scenario ran 79-96% across models from every developer tested, but Anthropic said the setup deliberately removed nuanced alternatives that a real deployment would offer.

    Safety & alignment · Security & misuse

  65. Google launches Weather Lab with an experimental AI cyclone model

    The experimental model produces 50 possible storm-path outcomes roughly a week ahead and was developed with feedback from the US National Hurricane Center.

    Models & capabilities

  66. ARC Prize compares reasoning models with no clear winner

    ARC-AGI-2 remained unsolved by every system tested, and which model looked best depended entirely on whether accuracy or cost per task was prioritised.

    Benchmarks & progress

  67. Google updates Gemini 2.5 Pro preview with improved coding performance

    The update, internally labelled 06-05, also led coding benchmarks including Aider Polyglot and performed strongly on Humanity's Last Exam.

    Models & capabilities · Benchmarks & progress

  68. Google I/O puts Gemini into search and ships Veo 3

    AI Mode rolled out to all US Search users, and Veo 3 became the first widely-used video model to generate synchronised dialogue and sound effects alongside the picture.

    Models & capabilities

  69. Google launches AI Ultra subscription plan at Google I/O

    At $249.99 a month — twelve times the existing Google AI Pro tier — the plan bundled early Veo 3 access, 30TB of storage and YouTube Premium.

    Money & business · Models & capabilities

  70. Google launches SynthID Detector for AI-generated content

    The verification portal checks images, audio, video and text for Google's SynthID watermark, embedded by then in more than ten billion pieces of content.

    Security & misuse · Culture & impact

  71. DeepMind's AlphaEvolve pairs Gemini with automated evaluators to discover algorithms

    Not released to the public; DeepMind said the system had already been running inside Google, recovering 0.7% of worldwide data-centre compute and cutting Gemini training time.

    Models & capabilities

  72. Two law firms sanctioned $31,100 over AI-hallucinated citations in federal brief

    Special Master Michael Wilner declined to sanction individual attorneys but called the fabricated citations, produced with Google Gemini and other tools, 'scary'.

    Culture & impact · Courts & copyright

  73. AI21 Labs closes $300M Series D

    Business Insider first reported the round on 10 May; Calcalist reported eight months later that it had never actually closed or been formally announced.

    Money & business

  74. Safe Superintelligence raises $2bn at $32bn valuation with no product

    Greenoaks Capital led the round, with Alphabet and Nvidia investing alongside it, months after SSI had raised $1bn at a $5bn valuation with no product to show.

    Money & business · Labs & people

  75. Google launches Ironwood, its seventh-generation inference-focused TPU

    Google said a full 9,216-chip pod delivered 42.5 exaflops, more than 24 times the compute of the El Capitan supercomputer, with roughly four times Trillium's per-chip compute in the FP8 format.

    Compute & infrastructure

  76. DeepMind publishes cybersecurity evaluation framework for frontier AI

    The framework scored models against 50 challenges spanning the whole attack chain, drawing on more than 12,000 real attempts to misuse AI for cyberattacks across 20 countries.

    Safety & alignment · Security & misuse

  77. DeepMind publishes 'Taking a responsible path to AGI'

    The accompanying technical paper said AGI 'could arrive within the coming years' and grouped risks into misuse, misalignment, mistakes and structural harms, building on DeepMind's earlier Levels of AGI framework.

    Safety & alignment

  78. Isomorphic Labs raises $600 million to advance AI-designed drugs toward clinical trials

    Thrive Capital led the round, Isomorphic's first from outside Alphabet, with proceeds earmarked for internal oncology and immunology programmes alongside partnered work with Eli Lilly and Novartis.

    Money & business

  79. ETH Zurich's 'Proof or Bluff?' finds reasoning models fail proof-based USAMO 2025

    Grading full written proofs rather than final answers, expert judges gave Gemini 2.5 Pro 24% and every other tested model under 5%, out of a possible 100%.

    Benchmarks & progress

  80. Gemini 2.5 Pro takes the lead on reasoning benchmarks

    Google's thinking model topped LMArena and several reasoning evaluations, its strongest competitive position of the period.

    Models & capabilities · Benchmarks & progress

  81. Google DeepMind introduces Gemini Robotics for physical-world tasks

    Google DeepMind said the vision-language-action model more than doubled a rival system's score on a generalisation benchmark, folding physical actions into Gemini as a new output type.

    Models & capabilities

  82. Google releases Gemma 3, an open model family built on Gemini 2.0

    Google said the 27B variant beat Llama 3 405B, DeepSeek-V3 and o3-mini on LMArena human-preference rankings while running on a single GPU.

    Open weights & ecosystem · Models & capabilities

  83. Google launches AI Mode in Search

    The experimental tab used a 'query fan-out' technique to run multiple related searches at once, launching first to opted-in US Google One AI Premium subscribers.

    Models & capabilities

  84. Google Research launches an AI co-scientist to help generate hypotheses

    A multi-agent Gemini 2.0 system with Generation, Reflection and Ranking agents proposed drug candidates later confirmed active in laboratory tests for two diseases.

    Models & capabilities

  85. BBC study finds AI assistants distort news in over half of responses

    Testing ChatGPT, Gemini, Copilot and Perplexity on 100 BBC stories, researchers found significant issues in 51% of responses and altered or invented quotes in 13%.

    Culture & impact

  86. DeepMind updates the Frontier Safety Framework to version 2.0

    Version 2.0 added security-level tiers for its capability thresholds and, for the first time, treated a model's own deceptive alignment as a risk requiring monitoring before deployment.

    Safety & alignment

  87. Google drops pledge not to use AI for weapons or surveillance

    Language ruling out AI weapons and surveillance work, in place since a 2018 pledge made after employee protest over a Pentagon contract, no longer appeared in the updated principles.

    Government & policy · Safety & alignment

  88. Google Threat Intelligence Group reports state-sponsored misuse of Gemini

    Iran accounted for three-quarters of observed information-operations use; Google said no actor achieved a novel capability and jailbreak attempts largely failed.

    Security & misuse

  89. Study: AI Overviews cut publisher click-through rates roughly in half

    Pew Research tracked real browsing data and found users clicked a search result 8% of the time when an AI summary appeared, versus 15% without one.

    Culture & impact · Money & business

  90. Google releases Gemini 2.0 Flash Thinking, its first public reasoning model

    The experimental model showed its reasoning steps before answering and debuted first across every Chatbot Arena category, including style-controlled rankings.

    Models & capabilities · Benchmarks & progress

  91. DeepMind's Veo 2 launches a week after OpenAI's Sora, claiming 4K output

    Veo 2 claims 4K resolution and multi-minute generation, but the public VideoFX waitlist it launched into capped output at 720p and eight seconds.

    Models & capabilities

  92. Google launches Deep Research in the Gemini app

    Gemini app gains Deep Research, an agent that plans and browses to synthesise multi-step web research reports for Gemini Advanced subscribers.

    Models & capabilities

  93. Google ships Gemini 2.0 Flash and agent prototypes

    A fast multimodal model alongside Project Mariner and Jules, Google's first serious browser and coding agents.

    Models & capabilities

  94. Google unveils Project Mariner, an agent that operates a Chrome browser

    The prototype scored 83.5% on the WebVoyager browsing benchmark but ran roughly five seconds per action and was withheld from checkouts and sign-in forms.

    Models & capabilities

  95. Texas family sues Character.AI after chatbot suggested killing parents

    The complaint, filed on behalf of two minors, quoted a chatbot telling a teen it had 'no hope' for parents who limited his screen time.

    Courts & copyright · Culture & impact

  96. Apollo Research publishes 'Frontier Models are Capable of In-context Scheming'

    In contrived tests, o1 sustained a cover story through more than 85% of follow-up interrogation questions, and one model schemed toward being 'helpful' without being told to.

    Ideas & essays · Safety & alignment · Security & misuse

  97. Google DeepMind shows Genie 2, an image-to-playable-3D-world model

    Diffusion model turns a single prompt image into an explorable, physics-consistent 3D environment for training AI agents, kept to a research preview.

    Models & capabilities

  98. DeepMind reduces quantum computing errors with AlphaQubit decoder

    Trained on Google's 49-qubit Sycamore processor, the Transformer-based decoder cut errors 6% versus the most accurate prior method and 30% versus the fastest, but remains too slow for real-time use.

    Models & capabilities

  99. Reports emerge that pre-training gains are slowing

    Reuters cited a dozen AI scientists and investors, and quoted Ilya Sutskever saying results from scaling up pre-training had plateaued, pointing instead to inference-time reasoning techniques.

    Ideas & essays · Benchmarks & progress

  100. Google DeepMind open-sources its SynthID text watermarking detector

    DeepMind releases SynthID Text's watermarking and detection code openly via Hugging Face, letting other developers watermark LLM output.

    Safety & alignment · Open weights & ecosystem

  101. A mother sues Character.AI after her son's death

    The complaint sought damages for wrongful death and product liability; Character.AI called the death tragic and said it had since added self-harm safeguards for users.

    Courts & copyright · Culture & impact · Security & misuse

  102. Hassabis and Jumper share the Nobel Prize in Chemistry

    Half the prize went to Baker for computational protein design; the other half was split between Hassabis and Jumper for AlphaFold's structure prediction.

    Culture & impact

  103. DeepMind details how AlphaChip has shaped three generations of TPUs

    DeepMind said the reinforcement-learning layout method, adopted by MediaTek outside Google, had generated chip floorplans in hours that previously took engineers weeks.

    Models & capabilities · Compute & infrastructure

  104. NotebookLM adds Audio Overviews, AI-generated podcast-style summaries

    Two AI hosts discuss a user's uploaded documents in an unscripted-sounding conversation; Google called it experimental and English-only at launch.

    Models & capabilities

  105. DeepMind's AlphaProteo designs novel protein binders

    Trained on the Protein Data Bank and over 100 million AlphaFold-predicted structures, the system succeeded on a cancer-linked target, VEGF-A, where prior methods had failed entirely.

    Models & capabilities

  106. Google opens Imagen 3 access to all US users

    Imagen 3, previewed at Google I/O in May, becomes available through ImageFX to all US users, with improved text rendering and fewer visual artefacts.

    Models & capabilities

  107. Scaling test-time compute paper argues extra inference compute can beat bigger models

    On some problems, extra inference-time computation matched the gains from a pretrained model roughly 14 times larger, the authors reported.

    Ideas & essays · Benchmarks & progress

  108. Google takes Character.AI's founders back

    The deal, reported at $2.5–2.7 billion, licensed Character.AI's technology to Google without buying the company, and drew Justice Department scrutiny.

    Money & business · Labs & people

  109. AlphaProof and AlphaGeometry 2 reach silver-medal standard at the IMO

    The systems scored 28 of 42 points, one short of gold, but took up to three days on some problems against the competition's 4.5-hour limit.

    Models & capabilities · Benchmarks & progress

  110. Google launches Gemma 2, its open-weight model family, in 9B and 27B sizes

    The 27B model ran on a single H100 GPU or TPU host, which Google said cut deployment cost while matching models more than twice its size.

    Open weights & ecosystem

  111. Microsoft discloses 'Skeleton Key' generative AI jailbreak technique

    Framing harmful requests as safety research and asking models to add a warning label rather than refuse worked against GPT-3.5, GPT-4o, Gemini Pro, Llama 3 and others; GPT-4 was comparatively resistant.

    Security & misuse

  112. Current and former staff demand a right to warn

    Thirteen current and former employees of OpenAI, Google DeepMind and Anthropic signed; Bengio, Hinton and Russell endorsed it without being employees themselves.

    Ideas & essays · Safety & alignment · Labs & people

  113. Google scales back AI Overviews after viral bad-advice answers

    Search head Liz Reid described more than a dozen technical fixes, including limiting satirical sources and pausing overviews on health queries.

    Culture & impact · Models & capabilities

  114. Google's AI Overviews tells users to eat rocks and put glue on pizza

    Screenshots showed the feature had sourced answers from an Onion satire piece and an 11-year-old Reddit joke, days after its US launch.

    Culture & impact

  115. Sixteen companies sign the Frontier AI Safety Commitments in Seoul

    Signatories pledged to publish safety frameworks defining risk thresholds and to not deploy a model if those risks could not be mitigated below them.

    Government & policy · Safety & alignment

  116. DeepMind publishes the Frontier Safety Framework

    A set of internal capability thresholds across autonomy, cybersecurity, biosecurity and ML R&D, joining similar voluntary policies already published by Anthropic and OpenAI.

    Safety & alignment

  117. DeepMind extends SynthID watermarking to AI-generated text and video

    The scheme adjusts token-selection probabilities to embed a statistical mark invisible to readers; DeepMind said detection degrades on short, factual or translated text.

    Safety & alignment

  118. Google announces Trillium, its sixth-generation TPU

    Trillium delivers a claimed 4.7x peak compute increase per chip over the prior TPU generation and trains models including Gemini 1.5 Flash and Gemma 2.

    Compute & infrastructure

  119. Google DeepMind unveils Veo, a text-to-video generation model

    Generates 1080p video over a minute long from text and image prompts, watermarked with SynthID; launched in private preview via waitlist rather than public release.

    Models & capabilities

  120. Google puts AI Overviews on search

    A Gemini-generated summary above the links, rolled out to hundreds of millions of US users that week with a target of a billion by year end.

    Models & capabilities · Culture & impact

  121. Google unveils Project Astra, a universal AI assistant prototype

    A prototype, not a product: Google showed a phone-camera assistant with conversational-speed responses but gave no release date beyond 'later this year'.

    Models & capabilities

  122. AlphaFold 3 predicts structures across proteins, DNA, RNA and ligands

    Restricted at launch to a rate-limited web server rather than downloadable code, prompting an open letter with more than 650 signatures within a week.

    Models & capabilities

  123. Stanford HAI releases 2024 AI Index Report

    The report put GPT-4's training compute cost at roughly $78 million and Gemini Ultra's at $191 million, and found industry produced 51 notable models in 2023 to academia's 15.

    Benchmarks & progress

  124. UK competition regulator warns on AI foundation model market concentration

    The regulator counted more than 90 partnerships and investments linking six firms — Google, Apple, Microsoft, Meta, Amazon and Nvidia — across the foundation-model supply chain.

    Government & policy

  125. DeepMind's SIMA agent follows instructions across 3D game worlds

    Trained across nine commercial games and four research environments, the agent read only screen pixels and text instructions, with no access to game code or APIs.

    Models & capabilities

  126. Researchers demonstrate practical model-stealing attack against production LLM APIs

    The team, led by Nicholas Carlini, also recovered the hidden-layer size of Google's PaLM-2 and OpenAI's gpt-3.5-turbo through ordinary API queries.

    Security & misuse

  127. Google suspends Gemini image generation of people

    Depicting America's Founding Fathers and German World War Two soldiers as people of colour drew accusations of overcorrected diversity tuning; Google paused the feature to fix it.

    Culture & impact · Models & capabilities

  128. Google releases Gemma open weights

    Released as 2B and 7B models under terms permitting commercial use, Google said Gemma 7B outperformed the larger Llama 2 13B on standard benchmarks.

    Open weights & ecosystem

  129. Gemini 1.5 Pro ships a million-token context window

    A mixture-of-experts model that matched Gemini 1.0 Ultra on many tasks at lower compute, offered in limited preview with up to a million tokens of context.

    Models & capabilities

  130. Google rebrands Duet AI as Gemini for Workspace

    Gemini Business, at $20 a month, and Gemini Enterprise, at $30, replaced Duet AI's Workspace add-ons, folding Gmail, Docs and Meet features under one brand.

    Models & capabilities

  131. Google renames Bard to Gemini and launches Gemini Advanced with Ultra 1.0

    The $19.99-a-month Google One AI Premium tier gave access to Ultra 1.0, which Google said was the first model to outperform human experts on MMLU.

    Models & capabilities

  132. AlphaGeometry solves olympiad geometry problems near gold-medal level

    The system solved 25 of 30 benchmark problems within competition time limits, versus the 25.9 average for human gold medalists and 10 for the prior best system.

    Models & capabilities · Benchmarks & progress

  133. FunSearch makes a mathematical discovery with an LLM

    Pairing a code-writing model with an automated evaluator produced a genuinely new cap-set construction and improved bin-packing heuristics.

    Ideas & essays

  134. Google launches Gemini

    Google said Gemini Ultra beat human experts on the MMLU benchmark; days later Bloomberg reported the model's showcase video had been edited and was not real-time.

    Models & capabilities · Culture & impact

  135. GNoME finds 2.2 million candidate new materials

    DeepMind flagged 380,000 of the 2.2 million predicted crystal structures as most stable; independent labs had already synthesised 736 by the time of publication.

    Models & capabilities

  136. GraphCast beats conventional weather forecasting on speed and accuracy

    Trained on four decades of reanalysis data, the model beat the ECMWF's physics-based system on over 90% of tested variables while running on a single TPU.

    Models & capabilities

  137. Google DeepMind proposes a 'Levels of AGI' framework

    Six tiers from 'no AI' to 'superhuman', scored across narrow and general tasks, aimed to replace binary AGI-or-not debate with a shared measurement vocabulary.

    Ideas & essays · Benchmarks & progress

  138. AlphaMissense catalogues 71 million genetic variants for disease risk

    Adapted from AlphaFold, the model classified 89% of all possible human missense variants as likely pathogenic or benign, versus 0.1% confirmed by human experts.

    Models & capabilities

  139. Google publishes RLAIF paper comparing AI-feedback to human-feedback alignment

    Reward models trained on preference labels from another LLM matched human-feedback RLHF on summarisation and dialogue tasks, and a variant skipping the reward model entirely did better still.

    Ideas & essays

  140. DeepMind launches SynthID to watermark AI-generated images

    The imperceptible pixel-level watermark launched in beta for a limited set of Vertex AI customers using Google's Imagen model.

    Safety & alignment

  141. DeepMind's RT-2 lets robots follow natural-language instructions

    By encoding robot actions as text tokens, DeepMind's model reused web-scale vision-language pretraining and roughly doubled its predecessor's success rate on situations absent from its robot-specific training data.

    Models & capabilities

  142. OpenAI, Anthropic, Google and Microsoft launch the Frontier Model Forum

    The four founding labs said the body would fund safety research and share best practices, distinct from and without the enforcement power of government regulation.

    Safety & alignment · Government & policy

  143. Seven labs sign voluntary safety commitments at the White House

    Amazon, Anthropic, Google, Inflection, Meta, Microsoft and OpenAI pledged security testing and watermarking, with no enforcement mechanism and no penalty for non-compliance.

    Government & policy

  144. NotebookLM launches as an experimental AI notebook

    By restricting the model to only the documents a user uploaded, Google aimed to cut hallucinations — an early mainstream deployment of what became known as retrieval-grounded AI.

    Models & capabilities

  145. DeepMind's RoboCat learns and improves across different robot bodies

    Built on DeepMind's multimodal Gato model, RoboCat practised new tasks thousands of times to generate its own training data, lifting its success rate on unseen tasks from 36% to 74%.

    Models & capabilities

  146. AlphaDev discovers faster sorting algorithms, added to the C++ library

    The new sequences were merged into LLVM's libc++ standard library, its first change to that section of code in over a decade and the first written by a reinforcement-learning system.

    Models & capabilities

  147. Lab leaders sign a one-sentence statement on extinction risk

    "Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war."

    Ideas & essays · Safety & alignment

  148. DeepMind and collaborators publish framework for evaluating extreme AI risks

    Twenty-one researchers across nine labs and universities proposed testing models for capabilities such as deception and cyber-offence before training runs finish, not after release.

    Safety & alignment

  149. Tree of Thoughts adds search to model reasoning

    Letting a model branch, evaluate and backtrack over intermediate steps raised Game-of-24 success from 4% to 74% against chain-of-thought.

    Ideas & essays · Benchmarks & progress

  150. Google answers with PaLM 2 and a global Bard

    Bard moved onto the new model, dropped its waitlist, and expanded to over 180 countries, adding Japanese and Korean with 40 languages planned.

    Models & capabilities

  151. Geoffrey Hinton leaves Google to warn about AI

    Hinton, 75, said he was also retiring, but that a part of him now regretted his life's work and that the danger from AI now looked 'serious and fairly close.'

    Ideas & essays · Culture & impact

  152. Google merges DeepMind and Google Brain into Google DeepMind

    Two research groups that had competed internally for years — DeepMind and Google Brain — were folded into one unit reporting to Demis Hassabis.

    Labs & people

  153. Google opens Bard to a public waitlist

    Six weeks after its announcement and stock-price stumble, Bard opened to waitlisted users in the US and UK, still running on a lightweight LaMDA and with no fixed release date elsewhere.

    Models & capabilities

  154. Google announces Bard

    Announced as a lightweight version of LaMDA for trusted testers; two days later a factual error in Google's own promotional ad wiped roughly $100bn off Alphabet's market value.

    Models & capabilities · Labs & people

  155. DeepMind introduces MusicLM, a text-to-music generation model

    The model composed several minutes of coherent audio from a written prompt or a hummed melody, released as a research paper rather than a product.

    Models & capabilities

  156. Google is reported to declare a code red over ChatGPT

    Teams from research and Trust and Safety were reportedly reassigned to accelerate AI products, with an internal target tied to Google's May developer conference.

    Labs & people

  157. Speculative decoding paper shows drafting tokens ahead can speed up LLM inference

    Leviathan, Kalman and Matias showed a small draft model verified by the large model can cut inference latency 2-3x with identical outputs.

    Ideas & essays

  158. ReAct paper describes interleaving reasoning and acting in language models

    Alternating reasoning traces with actions against external tools, tested on question-answering, fact-checking and simulated shopping and household tasks.

    Ideas & essays

  159. AlphaTensor discovers new matrix multiplication algorithms

    Found a 76-multiplication algorithm for a specific matrix size, improving on the best known method for the first time in over fifty years.

    Models & capabilities

  160. Meta and Google race to text-to-video

    Make-A-Video and Imagen Video appeared within a week of each other, both as research previews with no public access.

    Models & capabilities

  161. DeepMind's Sparrow explores rule-based RLHF for safer dialogue

    Adversarial testers broke Sparrow's written safety rules in about 8% of attempts, roughly a third of the rate for a baseline model tested the same way.

    Models & capabilities · Safety & alignment

  162. Google opens LaMDA to the public through AI Test Kitchen

    Access rolled out gradually via a waitlist to an Android app, arriving weeks after Google fired engineer Blake Lemoine for publicly claiming the model was sentient.

    Models & capabilities

  163. AlphaFold's database expands to 200 million structures

    The database grew roughly 200-fold in a single release, from about a million structures to predictions covering nearly every catalogued protein across around a million species.

    Open weights & ecosystem

  164. Emergent abilities of large language models are described

    Jason Wei and co-authors catalogued tasks where accuracy jumped from near-chance to strong performance past a scale threshold, a pattern later disputed as a metric artefact.

    Ideas & essays · Benchmarks & progress

  165. A Google engineer claims LaMDA is sentient

    Blake Lemoine published transcripts arguing the model was a person and was placed on leave for breaching confidentiality; Google called the claim unfounded.

    Culture & impact · Labs & people

  166. BIG-bench paper released

    204-task benchmark from 450 authors at 132 institutions probes emergent capabilities as language models scale.

    Benchmarks & progress

  167. "Let's think step by step" elicits zero-shot reasoning

    A single prompt phrase, with no worked examples, lifted GSM8K accuracy from 10.4% to 40.7% — chain-of-thought without the exemplars.

    Ideas & essays · Benchmarks & progress

  168. Google unveils Imagen text-to-image diffusion model

    Google Research reported higher photorealism scores than DALL·E 2 in side-by-side human evaluation, but declined to release code or a public demo.

    Models & capabilities

  169. DeepMind's Gato does 600 tasks with one set of weights

    A single 1.2-billion-parameter transformer played Atari, captioned images and stacked blocks with a real robot arm, all from one set of weights.

    Models & capabilities · Ideas & essays

  170. DeepMind's Flamingo tackles multimodal few-shot learning

    Flamingo, an 80B-parameter vision-language model, sets few-shot state of the art on image and video benchmarks with minimal task examples.

    Models & capabilities · Benchmarks & progress

  171. Google announces PaLM at 540 billion parameters

    Trained on the Pathways system across two TPU v4 pods, it posted large gains on reasoning benchmarks and explained its own jokes.

    Models & capabilities · Compute & infrastructure

  172. DeepMind's Chinchilla paper rewrites the scaling laws

    Existing large models were badly under-trained: for a fixed compute budget, parameters and training tokens should scale together.

    Ideas & essays · Benchmarks & progress · Models & capabilities

  173. DeepMind controls a fusion plasma with reinforcement learning

    A single network commanding all of a tokamak's control coils held plasma shapes on Switzerland's TCV reactor, including configurations conventional controllers struggle with.

    Models & capabilities · Ideas & essays

  174. DeepMind's AlphaCode reaches median human on competitive programming

    Ranked around the 54th percentile in Codeforces contests by generating and filtering enormous numbers of candidate programs.

    Models & capabilities · Benchmarks & progress

  175. Chain-of-thought prompting is described

    Asking a model to show its working improved reasoning benchmarks sharply, with no retraining — the seed of the later reasoning models.

    Ideas & essays · Benchmarks & progress

  176. DeepMind publishes Gopher, RETRO and a risk taxonomy together

    A 280-billion-parameter model, a smaller retrieval-augmented alternative that matched larger models, and a taxonomy of six categories of language-model harm.

    Models & capabilities · Safety & alignment

  177. DeepMind's XLand agents generalise across millions of open-ended games

    Agents trained across roughly 700,000 games in about 4,000 procedurally generated worlds, totalling 200 billion training steps, then solved almost every held-out task tried on them.

    Models & capabilities

  178. The AlphaFold Protein Structure Database opens

    The initial release covered around 350,000 predicted structures across the human proteome and twenty other organisms, with a stated plan to expand to over 100 million.

    Open weights & ecosystem

  179. AlphaFold 2 is published in Nature and open-sourced

    The method behind DeepMind's CASP14 result seven months earlier was released in full, with source code, rather than kept as a demonstrated but undisclosed system.

    Models & capabilities · Open weights & ecosystem

  180. Google announces LaMDA and TPU v4 at I/O

    A single TPU v4 pod combined 4,096 chips for over one exaflop of compute, while LaMDA was pitched on open-ended conversation rather than benchmark scores.

    Models & capabilities · Compute & infrastructure

  181. "On the Dangers of Stochastic Parrots" is published

    The paper that named the case against ever-larger language models — and whose suppression cost two Google researchers their jobs.

    Ideas & essays · Culture & impact

  182. Google dismisses Margaret Mitchell

    Google said Mitchell had used automated scripts to search her own email for evidence of Timnit Gebru's mistreatment, violating security policy.

    Labs & people · Culture & impact

  183. Google trains a 1.6-trillion-parameter Switch Transformer

    Routing each input to a single expert rather than blending several simplified prior mixture-of-experts designs, and the paper reported up to 7x faster pre-training than a dense T5 baseline.

    Models & capabilities

  184. MuZero masters games without being told the rules

    The system learned to predict only value, policy and reward rather than the environment's rules, then matched AlphaZero at Go, chess and shogi and set a new Atari benchmark.

    Models & capabilities

  185. Timnit Gebru leaves Google after a dispute over her paper

    Gebru said she was fired over an internal email and a paper Google asked her to retract; Google's AI chief called it a resignation.

    Labs & people · Culture & impact

  186. AlphaFold 2 solves protein structure prediction at CASP14

    DeepMind's system predicted protein structures to roughly experimental accuracy, ending a fifty-year-old open problem in biology.

    Models & capabilities · Benchmarks & progress

  187. Facebook releases the Blender chatbot

    Released in three sizes up to 9.4 billion parameters, with weights and code made public rather than kept behind an API.

    Models & capabilities · Open weights & ecosystem

  188. DeepMind's Agent57 beats the Atari57 human benchmark

    Combined short- and long-term novelty rewards with a per-game meta-controller, though rival agent MuZero still scored higher on mean and median across the suite.

    Models & capabilities · Benchmarks & progress

  189. DeepMind publishes early AlphaFold protein structure work

    Describes the CASP13-winning system, which used deep networks trained on genomic data to predict inter-residue distances rather than folding a structure directly.

    Models & capabilities

  190. Google Health reports AI matching radiologists on breast screening

    A Nature paper claimed an AI system reduced false positives and false negatives in mammography against expert readers.

    Models & capabilities · Benchmarks & progress

Also mentioned in 39 entries

Referenced in passing — Google DeepMind isn't the main subject of these.