Verity Harding
@verityharding.bsky.social
Director, AI & Geopolitics, Uni of Cambridge | Founder, Formation Advisory | TIME100 AI | Former DeepMind, Google, UK govt | King’s Trust | Author of “AI Needs You: how we can change AI’s future and save our own” https://tinyurl.com/y4v26spa
In the words of the great @tricialockwood.bsky.social : no one is talking about this. My latest piece on the growing support for open source models and why the OpenAI hack on Hugging Face may have sped up an underlying trend most haven’t noticed yet.
Inflection Points
Once in a while, everything shifts.
verityharding.substack.com
On what counts as a true inflection point in AI, and why OpenAI’s Hugging Face hack is one, but not for the reasons you think. My latest post for the AI Needs You substack. Free to read and subscribe.
Inflection Points
Once in a while, everything shifts.
verityharding.substack.com
On what counts as a true inflection point in AI, and why OpenAI’s Hugging Face hack is one, but not for the reasons you think. My latest post for the AI Needs You substack. Free to read and subscribe.
Inflection Points
Once in a while, everything shifts.
verityharding.substack.com
Chris Nolan vs George Lucas — who is right about the future of AI in Hollywood?
Nolan vs Lucas: Hollywood's AI battle
Creators are split on AI in filmmaking. The audience is picking sides.
open.substack.com
Chris Nolan vs George Lucas — who is right about the future of AI in Hollywood?
Nolan vs Lucas: Hollywood's AI battle
Creators are split on AI in filmmaking. The audience is picking sides.
open.substack.com
Chris Nolan vs George Lucas: who’s right about the future of AI in Hollywood? My latest post out now:
Verity Harding (@verityharding)
Chris Nolan vs George Lucas: who’s right about the future of AI in Hollywood? Latest post:
substack.com
The framing of AI as an “arms race” between the US and China is not just wrong, it’s dangerous. My new anthology, "Reimagining the AI Arms Race" features contributions from a range of expert leaders all interrogating the accuracy and relevance of that framing. Here’s a preview of 5 of them:
The framing of AI as an “arms race” between the US and China is not just wrong, it’s dangerous. My new anthology, "Reimagining the AI Arms Race" features contributions from a range of expert leaders all interrogating the accuracy and relevance of that framing. Here’s a preview of 5 of them:
🚨New publication 🚨 Today my group at Cambridge, the AI & Geopolitics Project, launches our first anthology. It’s a bold new collection of essays bringing together 10 thought-leaders to challenge the dominant narrative that AI is an existential, zero-sum battle between the US and China (1/3)
🚨New publication 🚨 Today my group at Cambridge, the AI & Geopolitics Project, launches our first anthology. It’s a bold new collection of essays bringing together 10 thought-leaders to challenge the dominant narrative that AI is an existential, zero-sum battle between the US and China (1/3)
If you want to understand what happened with Anthropic, Mythos and the US government this weekend, you need to understand the politics of AI in D.C. right now. My latest post is live:
What’s going on in D.C.?
Inside the civil wars reshaping US AI policy
open.substack.com
If you want to understand what happened with Anthropic, Mythos and the US government this weekend, you need to understand the politics of AI in D.C. right now. My latest post is live:
What’s going on in D.C.?
Inside the civil wars reshaping US AI policy
open.substack.com
AI is now so unpopular amongst Democrat core voters that in the LA mayoral race Karen Bass clearly feels “his campaign uses AI” is an effective enough attack line—even in such a short rebuttal as this (from the @nytimes.com)
AI is now so unpopular amongst Democrat core voters that in the LA mayoral race Karen Bass clearly feels “his campaign uses AI” is an effective enough attack line—even in such a short rebuttal as this (from the @nytimes.com)
For a while it looked like a truce, but Republicans’ AI fault lines have reemerged. Last week conservative influencers including Bannon urged Trump to make pre-deployment testing of frontier models mandatory. Days later, Trump shelved an executive order on AI safety after industry lobbying (1/2)
AI companies are writing their own constitutions. Meanwhile, a shared vocabulary for contesting algorithmic power has been traveling across red and blue states, from legislatures to civil society. New from me in @science.org: "A Civic Grammar for AI Rights" www.science.org/doi/10.1126/... 🧵
A civic grammar for AI rights
Earlier this year, the artificial intelligence (AI) company Anthropic, developer of the large language model Claude, published a document it calls “Claude’s Constitution,” described by the company as ...
science.org
An interesting dynamic in the AI jobs apocalypse discourse is how everyone thinks that their area of expertise is safe, but that areas they don’t know much about are at risk. Eg writers being able to recognise AI writing as turgid and unoriginal; SWEs seeing vibe coding as buggy and unsafe… (1/3)
Read my last two posts to understand (some of) what’s going on here: open.substack.com/pub/verityha... open.substack.com/pub/verityha...
Does AI have a 'cool' problem?
For young people and artists, AI and its architects are the object of derision. Does it matter?
open.substack.com
404 Media@404media.co · 5mo agoFormer Google CEO Eric Schmidt was booed throughout this commencement speech at the University of Arizona for his praise of AI. This comes just a week after another commencement speaker who mentioned AI was booed at a school in Florida. Read more: www.404media.co/ucf-ai-comme...
Anna Wintour taken aback by the backlash to the Bezos Met Gala sponsorship. A commencement speaker in Florida shocked by the students booing her every mention of AI. Old people do not understand what’s happening with AI and the young. Here’s what’s going on, why it’s happening, and what it means:
Does AI have a 'cool' problem?
For young people and artists, AI and its architects are the object of derision. Does it matter?
open.substack.com
Anna Wintour taken aback by the backlash to the Bezos Met Gala sponsorship. A commencement speaker in Florida shocked by the students booing her every mention of AI. Old people do not understand what’s happening with AI and the young. Here’s what’s going on, why it’s happening, and what it means:
Does AI have a 'cool' problem?
For young people and artists, AI and its architects are the object of derision. Does it matter?
open.substack.com
Among Gen Z AI’s net favourability is now minus 44, driven by fears about jobs, the environment, creativity. But there’s another trend that many have missed: AI just isn’t very cool. My latest post explores the trend, the reasons behind it, and the consequences for the OpenAI/Anthropic rivalry:
Does AI have a 'cool' problem?
For young people and artists, AI and its architects are the object of derision. Does it matter?
open.substack.com
Among Gen Z AI’s net favourability is now minus 44, driven by fears about jobs, the environment, creativity. But there’s another trend that many have missed: AI just isn’t very cool. My latest post explores the trend, the reasons behind it, and the consequences for the OpenAI/Anthropic rivalry:
Does AI have a 'cool' problem?
For young people and artists, AI and its architects are the object of derision. Does it matter?
open.substack.com
The space race shows is that competition and collaboration needn't be mutually exclusive. I'd be amazed if, in 2026, a longer-term dialogue on safety wasn't be part of the agenda this week. Any hint of recognition that AI isn't just a race to be won would be a huge step in the right direction
In 1967 the UN agreed a treaty which proclaimed space as "the province of all mankind" and prohibited weapons of mass destruction from orbit and celestial bodies. It meant that, in 1969, "We Came in Peace for All Mankind" was not just a slogan but a statement underpinned by international law.
In 1967 the UN agreed a treaty which proclaimed space as "the province of all mankind" and prohibited weapons of mass destruction from orbit and celestial bodies. It meant that, in 1969, "We Came in Peace for All Mankind" was not just a slogan but a statement underpinned by international law.
Two years ago I wrote about what a missed opportunity the Biden-Xi Summit was for progress on AI cooperation. If only some foundations had been laid then, we would be in a lot better position today. All the same lessons from then still apply today:
What the cold war space race can teach us on AI
International agreement could curb the fight for supremacy in today’s most ambitious technology
ft.com
Two years ago I wrote about what a missed opportunity the Biden-Xi Summit was for progress on AI cooperation. If only some foundations had been laid then, we would be in a lot better position today. All the same lessons from then still apply today:
What the cold war space race can teach us on AI
International agreement could curb the fight for supremacy in today’s most ambitious technology
ft.com
At the Truth Tellers Summit today hosted by the indomitable @tinabrownlm.bsky.social Much discussion already of the challenges facing journalism in a world of unreality. Read our latest post on how Iran is serving as a “toxic laboratory” for the future of epistemic collapse
Deepfakes and the battle for reality in Iran
The country “has become a laboratory for some of the most toxic excesses” of disinformation, says Mahsa Alimardani, but there is a way forward.
open.substack.com
In this piece Mahsa talks clearly about the current information ecosystem in Iran. A lot of the lessons are equally applicable to misinformation in times of crisis as we’re seeing with the conspiracy theories about the assassination attempt. Recommend this interview to understand what’s happening.
Verity Harding@verityharding.bsky.social · 6mo agoI highly recommend this excellent interview between my colleague Catherine Westall and the world leading expert on deepfake and Iran. This week’s edition of AI Needs You:
In this piece Mahsa talks clearly about the current information ecosystem in Iran. A lot of the lessons are equally applicable to misinformation in times of crisis as we’re seeing with the conspiracy theories about the assassination attempt. Recommend this interview to understand what’s happening.
Verity Harding@verityharding.bsky.social · 6mo agoI highly recommend this excellent interview between my colleague Catherine Westall and the world leading expert on deepfake and Iran. This week’s edition of AI Needs You: