/
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

AI pioneer Yoshua Bengio launches LawZero, a nonprofit focused on safer AI; LawZero has raised nearly $30M from donors including Skype co-founder Jaan Tallinn

Turing Award winner launches LawZero not-profit for safer AI, while warning recent models display dangerous characteristics

Financial Times Cristina Criddle

Context & Ripple Effects

LawZero turns Yoshua Bengio’s long-running calls for public protection and AI regulation into a dedicated institution. His earlier support for California’s AI safety bill and warnings that governments should act quickly provide the policy backdrop.

The launch also creates an organizational home for a research agenda later explored in a LawZero discussion of safety research and autonomous systems, rather than leaving Bengio’s intervention solely in public advocacy.

First-order effects

  • LawZero gains an initial donor-funded base—nearly $30 million—to operate as a nonprofit focused on safer AI, giving Bengio a vehicle separate from commercial model developers.
  • The launch concentrates Bengio’s warnings about dangerous model characteristics into an institution that can produce safety research and public-facing analysis.

Second-order effects

  • The nonprofit adds an independent, well-funded voice to debates over model safeguards and regulation, reinforcing arguments Bengio had already advanced around urgent oversight of increasingly capable models.
  • Safety researchers, policymakers and donors gain another focal point outside the major AI labs, potentially broadening the sources of evidence and advocacy shaping governance debates.

Third-order effects

  • If similar efforts attract sustained backing, AI safety may become more institutionally divided between commercial labs, governments and independent nonprofit research groups—each with distinct incentives and claims to legitimacy.
  • The move is one data point in a shift toward public-safety governance of advanced AI, where independent institutions may increasingly seek to influence standards before capabilities become more autonomous.

The trend: AI safety is moving from individual researchers’ warnings toward dedicated, independently funded institutions that aim to shape how advanced-model risks are studied and governed.

Discussion

  • @eicathomefinn Margot Finn on bluesky
    Remind me again why advocates (including many senior managers and politicians) are exhorting school teachers and university lecturers to embed AI into every aspect of our curriculums?
  • @jjaron Jacob Aron on bluesky
    Sigh.  LLMs don't lie, because they don't know anything.  What all LLMs do, including the latest models, is produce statistically plausible outputs that sometimes resemble the truth and sometimes resemble falsehoods.  That's all the tech *can* do www.ft.com/content/2b3c...
  • @winwood66 Stevie Winwood on x
    “my concern is that any time in the future, the next version might be strategically intelligent enough to see us coming from far away and defeat us with deceptions that we don't anticipate. So I think we're playing with fire right now.” https://www.ft.com/...
  • @starcourse Nicholas Beale on x
    'Responsible AI is vitally important. Investors need to focus on it - there are grave systemic risks looming. Also pay attention to the Unethical Optimization Principle. https://www.ft.com/... via @ft
  • @dineshgovender Dinesh Govender on x
    “If we build AIs that are smarter than us and are not aligned with us and compete with us, then we're basically cooked.” ‘Godfather’ of AI Yoshua Bengio says latest models lie to users https://www.ft.com/... via @ft