/
Navigation
Chronicles
Browse all articles
Explore
Semantic exploration
Research
Entity momentum
Nexus
Correlations & relationships
Story Arc
Topic evolution
Drift Map
Semantic trajectory animation
Posts
Analysis & commentary
Pulse API
Tech news intelligence API
Browse
Entities
Companies, people, products, technologies
Domains
Browse by publication source
Handles
Browse by social media handle
Detection
Concept Search
Semantic similarity search
High Impact Stories
Top coverage by position
Sentiment Analysis
Positive/negative coverage
Anomaly Detection
Unusual coverage patterns
Analysis
Rivalry Report
Compare two entities head-to-head
Semantic Pivots
Narrative discontinuities
Crisis Response
Event recovery patterns
Connected
Search: /
Command: ⌘K
Embeddings: large
TEXXR

Chronicles

The story behind the story

days · browse · Enter similar · o open

A profile of, and interview with, DeepMind CEO Demis Hassabis on AGI, “not moving fast and breaking things”, launching the Sparrow chatbot in 2023, and more

Demis Hassabis stands halfway up a spiral staircase, surveying the cathedral he built. Tweets: @glenngabe , @billyperrigo , @paulroetzer , @time , @eileentso , @jeffnesbit , @sarthakgh , @deepmind , and @dylanmatt Tweets: Glenn Gabe / @glenngabe : Google's AI ambitions -> An interview with DeepMind's CEO About AI “He says DeepMind might release its own chatbot, Sparrow, for a private beta in 2023. The delay is to work on reinforcement learning-based features that ChatGPT lacks, like citing sources” https://time.com/... https://twitter.com/... Billy Perrigo / @billyperrigo : New: In a rare interview, DeepMind CEO Demis Hassabis tells me he is concerned about the speed at which AI is progressing, and urges his competitors to focus more on safety. “I would advocate not moving fast and breaking things,” he says. https://time.com/... Paul Roetzer / @paulroetzer : DeepMind is also considering releasing its own chatbot, called Sparrow, for “private beta” some time in 2023. (The delay is in order for DeepMind to work on reinforcement learning-based features that ChatGPT lacks, like citing its sources. https://twitter.com/... @time : With AI's promise also comes peril. It is in this uncertain climate that @demishassabis agrees to a rare interview with @billyperrigo, to issue a stark warning about his growing concerns https://time.com/... Eileen Burbidge / @eileentso : “@demishassabis was thoughtful enough to understand that the technology had long-term societal [ethical] implications, and wanted to understand those before the technology was invented, not after deployed,”... not just 2 steps ahead, but 20 steps ahead" https://time.com/... Jeff Nesbit / @jeffnesbit : “He likes the ideas of universal basic income, under which every citizen is given a monthly stipend from the government, and universal basic services, where the state pays for basic living standards like transportation or housing.” https://time.com/... Sar Haribhakti / @sarthakgh : Elon Musk is not only one of the cofounders of OpenAI but was also the earliest backer of now Google-owned Deep Mind which will likely launch its ChatGPT rival soon https://time.com/... https://twitter.com/... @deepmind : Our founder and CEO @DemisHassabis talks to @TIME about DeepMind's mission to build AI responsibly to help solve some of the world's toughest problems. 🌐 https://dpmd.ai/... Dylan Matthews / @dylanmatt : DeepMind turned down an acquisition offer from Facebook in favor of one for less money from Google because Facebook wouldn't agree to their ethical stipulations https://time.com/... https://twitter.com/...

TIME Billy Perrigo

Context & Ripple Effects

This TIME profile lands at the moment ChatGPT has reset public expectations for chatbots, and Hassabis is positioning DeepMind as the deliberate counterweight: he tells Billy Perrigo the lab may release its Sparrow chatbot as a private beta in 2023, delayed specifically to build reinforcement-learning features ChatGPT lacks, like citing sources. The 'not moving fast and breaking things' framing is consistent with how DeepMind has always operated — it turned down Facebook's competing offer to accept Google's acquisition over ethical stipulations.

The tension this interview surfaces — research patience versus product urgency inside Alphabet — becomes the throughline of subsequent coverage: by mid-2023 Hassabis is discussing DeepMind's restructuring and an AGI timeline alongside Google Bard, and by 2024, freshly knighted, he is still being asked whether he can navigate Google's bureaucracy and turn research into products (Big Technology's post-knighthood profile).

First-order effects

  • Sparrow enters a consumer chatbot market defined by ChatGPT, and DeepMind's stated differentiation — reinforcement-learning features such as source citations — makes verification, not raw fluency, its launch pitch.
  • Hassabis publicly stakes DeepMind's brand against the 'move fast and break things' ethos, urging competitors to prioritize safety while OpenAI's Musk-cofounded rival sets the pace.

Second-order effects

  • Alphabet now fields two conversational-AI efforts with different tempos — Bard shipping fast under Google, Sparrow deliberately under DeepMind — forcing internal questions about which culture governs Google's AI products, the friction executives later describe when rapidly infusing Google products with AI.
  • If source-citing chatbots prove commercially viable, competitors face pressure to add verifiable-citation features of their own, shifting differentiation from model scale toward trust and provenance.

Third-order effects

  • Hassabis's posture — high-minded goals over short-term revenue, a stance sources later say extended to pursuing a Nobel rather than maximizing Alphabet's race position (Reuters reporting) — models the 'state-compatible' lab: one built to be regulated, funded, and trusted by governments rather than to outrun them.
  • If the pattern holds, frontier labs bifurcate into speed-first and safety-first camps, with safety positioning becoming a durable competitive asset as regulators weigh dual-use AI governance.

The trend: Frontier AI labs are splitting into speed-first and deliberation-first camps, with DeepMind betting that safety-aligned, source-verifiable products will outlast the move-fast era.

Discussion

  • @paulroetzer Paul Roetzer on x
    DeepMind is also considering releasing its own chatbot, called Sparrow, for “private beta” some time in 2023. (The delay is in order for DeepMind to work on reinforcement learning-based features that ChatGPT lacks, like citing its sources. https://twitter.com/...
  • @time @time on x
    With AI's promise also comes peril. It is in this uncertain climate that @demishassabis agrees to a rare interview with @billyperrigo, to issue a stark warning about his growing concerns https://time.com/...
  • @eileentso Eileen Burbidge on x
    “@demishassabis was thoughtful enough to understand that the technology had long-term societal [ethical] implications, and wanted to understand those before the technology was invented, not after deployed,”... not just 2 steps ahead, but 20 steps ahead" https://time.com/...
  • @jeffnesbit Jeff Nesbit on x
    “He likes the ideas of universal basic income, under which every citizen is given a monthly stipend from the government, and universal basic services, where the state pays for basic living standards like transportation or housing.” https://time.com/...
  • @sarthakgh Sar Haribhakti on x
    Elon Musk is not only one of the cofounders of OpenAI but was also the earliest backer of now Google-owned Deep Mind which will likely launch its ChatGPT rival soon https://time.com/... https://twitter.com/...
  • @deepmind @deepmind on x
    Our founder and CEO @DemisHassabis talks to @TIME about DeepMind's mission to build AI responsibly to help solve some of the world's toughest problems. 🌐 https://dpmd.ai/...
  • @dylanmatt Dylan Matthews on x
    DeepMind turned down an acquisition offer from Facebook in favor of one for less money from Google because Facebook wouldn't agree to their ethical stipulations https://time.com/... https://twitter.com/...