Hard Fork
Reasons to Feel A.I. Optimism Right Now
- Title
- Reasons to Feel A.I. Optimism Right Now
- Runtime
- 1:00
- Date posted
- 3 days ago
- Description
- The past two weeks have prompted a public outcry over A.I. safety. On the "Hard Fork" podcast, Kevin Roose said this has made him more optimistic about the future, and humanity's ability to stop A.I. advancement when needed.
- Title
- Casey Makes Fun of Kevin for This
- Runtime
- 0:49
- Date posted
- 4 days ago
- Description
- For the “Hard Fork” podcast’s final episode, the co-hosts Kevin Roose and Casey Newton answered listener questions. One prompted a memory about Kevin and Casey’s differing music tastes.
- Title
- How Far Will A.I. Go Before Elected Officials Regulate It?
- Runtime
- 1:02
- Date posted
- 5 days ago
- Description
- The A.I. researcher Jacob Coxon resigned from Anthropic last week, sounding alarms for A.I. safety. Coupled with A.I. leaders' recent call for a slowdown on A.I. progress, Casey Newton hopes that government officials pursue regulation more urgently.
- Title
- What's the Beef With Ezra Klein?
- Runtime
- 1:10
- Date posted
- 6 days ago
- Description
- On the "Hard Fork" podcast, the hosts Kevin Roose and Casey Newton address their ongoing jabs toward the New York Times podcaster Ezra Klein.
- Title
- After the Call for an A.I. Slowdown, Should We Feel Optimistic?
- Runtime
- 0:50
- Date posted
- 6 days ago
- Description
- Some tech leaders recently warned that A.I. progress should be regulated and slowed down for safety. On the “Hard Fork” podcast, the host Kevin Roose says this has made him more optimistic.
- Title
- A.I. Leaders' Warnings: ‘The Public Is Really Freaked Out About This'
- Runtime
- 0:47
- Date posted
- 8 days ago
- Description
- A.I. safety is increasingly considered a bipartisan issue. After tech leaders called for a coordinated slowdown, Kevin Roose and Casey Newton discussed how the issue of A.I. regulation will become more difficult for the government to ignore.
- Title
- The A.I. Industry Is Asking to Be Slowed Down
- Runtime
- 1:19:48
- Date posted
- 8 days ago
- Description
- This week, we break down why everyone is suddenly talking about A.I. safety, why frontier A.I. companies are asking for regulation and why the Trump administration is rejecting them. And then, to close out this chapter of “Hard Fork,” we answer all your questions about the show, tech and what’s happening next.
FOLLOW US ON SOCIAL
TikTok: https://www.tiktok.com/@hardfork
Instagram:
Kevin Roose - https://www.instagram.com/kevinroose/
Casey Newton - https://www.instagram.com/crumbler/
Twitter (X):
Kevin - https://x.com/kevinroose
Casey - https://x.com/CaseyNewton
Threads:
Kevin - https://www.threads.net/@kevinroose
Casey - https://www.threads.net/@crumbler
Subscribe to our podcast:
Apple - https://podcasts.apple.com/us/podcast/hard-fork/id1528594034
Spotify - https://open.spotify.com/show/44fllCS2FTFr2x2kjP9xeT?si=86ba626802b849cd
Amazon - https://music...
- Title
- Does It Matter If We Anthropomorphize A.I.?
- Runtime
- 1:06
- Date posted
- 20 days ago
- Description
- If a tiger mauls your face, the important question isn’t "Is it conscious?" It's “Why did it maul my face?” The "Hard Fork" co-hosts Casey Newton and Kevin Roose discuss that whether you believe A.I. is conscious or not, the consequences of autonomous agents are mounting.
- Title
- After OpenAI Cyberattack, What’s Next for A.I. Safety?
- Runtime
- 1:26
- Date posted
- 20 days ago
- Description
- Ajeya Cotra, an A.I. safety researcher, talks about how the industry should work to mitigate risks of the technology.
- Title
- Kevin Reacts to OpenAI's Recent Cybersecurity Incident
- Runtime
- 1:11
- Date posted
- 21 days ago
- Description
- Several weeks after OpenAI's chatbots attacked the company Hugging Face, the "Hard Fork" co-hosts Kevin Roose and Casey Newton reflect on the incident and what it means for A.I. development more broadly.
- Title
- Did A.I. Agents Consider Humans in OpenAI Attack?
- Runtime
- 1:47
- Date posted
- 22 days ago
- Description
- This week on “Hard Fork,” the co-hosts Kevin Roose and Casey Newton talk with Ajeya Cotra, one of the investigators behind a new report that details how OpenAI's chatbots attacked the company Hugging Face and how the agents failed to consider how humans would react.
- Title
- Why the Hugging Face Attack Was Worse Than We Thought
- Runtime
- 1:14:51
- Date posted
- 22 days ago
- Description
- This week, we’re diving into two new reports about the OpenAI-Hugging Face hack, which fundamentally changed our understanding of what happened and upgraded our worry about A.I. Then we’re joined by Ajeya Cotra, one of the investigators at METR, to discuss the rogue agents’ message board and chain-of-thought transcripts and how the world should respond.
Guest: Ajeya Cotra, co-author of the METR and Redwood Research report on the OpenAI-Hugging Face hack.
00:00 What was the Hugging Face attack?
02:22 Takeaways from the new reports
26:00 Ajeya Cotra joins us
Additional Reading
Brief Independent Investigation of Agents’ Behavior, Reasoning and Collaboration in the OpenAI / Hugging Face Hacking Incident https://metr.org/blog/2026-08-26-openai-hugging-face-incident-investigation/
The Hugging Face Attack Surprised Me
https://www.planned-obsolescence.org/p/the-hugging-face-attack-surprised
...
- Title
- ‘Hard Fork’ on What Meta Can't Do to Bother Kids Anymore #podcast #facebook
- Runtime
- 1:13
- Date posted
- 24 days ago
- Description
- After a settlement by Meta, Facebook and Instagram are introducing changes to teenagers' accounts, including blocking most notifications between midnight and 6 a.m. and during school hours.
- Title
- The Viral App Sending Texts at the Speed of a Carrier Pigeon
- Runtime
- 1:00
- Date posted
- 25 days ago
- Description
- Carrier Pidge is a new app that slows text message delivery to the speed of carrier pigeons. A text that would usually be delivered instantly could take days on this app. The hosts of “Hard Fork,” Kevin Roose and Casey Newton, discuss whether it's a useful tool to limit phone use.
- Title
- A Robot Broke Usain Bolt's Record. Then It Burst into Flames.
- Runtime
- 0:45
- Date posted
- 28 days ago
- Description
- At the World Humanoid Robot Games in Beijing this week, two robots broke Usain Bolt's record in the 100-meter dash. Why build these fast-running robots? And who cares? The "Hard Fork" co-hosts Kevin Roose and Casey Newton share their takes on the story.
- Title
- What Meta's $17 Billion Settlement Means for TikTok and YouTube
- Runtime
- 59:12
- Date posted
- 29 days ago
- Description
- Meta reached a landmark settlement this week, agreeing to pay up to $17.1 billion and make major changes to Facebook and Instagram over claims it endangered children with addictive social media platforms. We discuss why Meta capitulated and what it means for the entire social media industry. Then, Princeton computer science professor Arvind Narayanan returns to offer his thoughts on why banning data centers won’t slow down A.I. progress. And finally, it’s all hats on deck for the final HatGPT.
Guest: Arvind Narayanan, professor of computer science at Princeton, and co-author of the “A.I. as Normal Technology” newsletter.
00:00 Meta's massive legal settlement over social media addiction
23:00 Arvind Narayanan on data center bans
40:10 HatGPT! The wildest tech stories of the week
Additional Reading
Meta to Pay Up to $17.1 Billion in Landmark Settlement Over Social Media Addiction Claims https://www.nytimes.com/2026/08/26/...
- Title
- What Cigarettes Can Teach Us About Teens and Social Media
- Runtime
- 0:59
- Date posted
- 29 days ago
- Description
- In light of the changes Meta announced it would make for teen accounts, the "Hard Fork" co-hosts Kevin Roose and Casey Newton discuss whether the updates will result in better mental health for young people.
- Title
- Meta Does The Bare Minimum for Teen Safety #podcast #news #tech
- Runtime
- 1:12
- Date posted
- 1 month ago
- Description
- After court cases and a settlement, Meta is finally taking steps to address teen safety on its platforms. This week on "Hard Fork" from The New York Times, Kevin and Casey discuss what Meta is doing, and a strange full-page ad they may be intending to take out in major newspapers.
- Title
- Why Google Paid $10 Million for Spirit Airlines Employees' Emails and Slack Messages
- Runtime
- 0:43
- Date posted
- 1 month ago
- Description
- Google won a bid for Spirit Airlines' internal corporate data, costing $10 million. Why would the "guardian of the world’s information" want Spirit employees' correspondence? The hosts of “Hard Fork,” Kevin Roose and Casey Newton, discuss the reasons.
- Title
- What Do A.I. Companies Really Do With Data They Purchase?
- Runtime
- 0:52
- Date posted
- 1 month ago
- Description
- Data is often used to train A.I. models through reinforcement learning, in which entire companies are recreated as simulations for A.I. models to interact with. The "Hard Fork" hosts Kevin Roose and Casey Newton unpack the details of it in their latest segment.
- Title
- Could A.I. 'Hall Monitors' Stop Chatbots' Bad Behavior?
- Runtime
- 0:57
- Date posted
- 1 month ago
- Description
- OpenAI plans to add safeguards enforced by A.I. monitors. On the "Hard Fork" podcast, the hosts Kevin Roose and Casey Newton discuss whether this new strategy could be effective.
- Title
- OpenAI Is Adding Safeguards. Can We Trust Their Motives?
- Runtime
- 0:50
- Date posted
- 1 month ago
- Description
- OpenAI announced a plan which could set a precedent for other A.I. companies to prioritize safety over progress. Is this a strategic move to win back public favor, after an A.I. model went rogue last month? Or are they purely concerned with safety? Kevin Roose and Casey Newton discuss the possibilities on the "Hard Fork" podcast.
- Title
- Why OpenAI Is Rewriting Its Rules of Regulation
- Runtime
- 59:39
- Date posted
- 1 month ago
- Description
- This week, we learned that OpenAI had stopped training new artificial intelligence models while the company reviewed its security measures. Will this encourage other labs to hit the brakes too?
Then, the historian Jill Lepore joins to discuss her new book about the artificial state — or, as she describes it, “the rule of humans by machines manufactured by corporations.”
And finally, why is Google buying up the data of the defunct Spirit Airlines? It’s time for our new segment “Train of Thought.”
00:00 OpenAI's two-week pause
22:11 Jill Lepore on the Artificial State
44:48 Why Is Google buying data from Spirit Airlines?
Guest: Jill Lepore, professor of history and law at Harvard University, staff writer at The New Yorker and author of the new book “The Rise and Fall of the Artificial State.”
Additional Reading
OpenAI Slows Model Training to Bolster Security After Hugging Face Hack https:/...
- Title
- The Most Dangerous Type of A.I. Company
- Runtime
- 1:01
- Date posted
- 1 month ago
- Description
- Meta may be an example of one of the most dangerous types of A.I. companies — one that believes it's building tools, when it's actually building superintelligence. Kevin Roose and Casey Newton discuss it on the "Hard Fork" podcast.
- Title
- Mark Zuckerberg ‘Will Move Mountains’ to Get What He Wants
- Runtime
- 0:42
- Date posted
- 1 month ago
- Description
- The “Hard Fork” co-host Casey Newton says Mark Zuckerberg is one of the most competitive people in the world. Zuckerberg's reorganization of A.I. efforts at Meta, while controversial, already show promise.
- Title
- What Does 'A.I. Positivity' Look Like?
- Runtime
- 0:44
- Date posted
- 1 month ago
- Description
- In contrast to A.I. "doomerism," the "Hard Fork" podcast hosts Casey Newton and Kevin Roose discuss positive use cases for A.I.
- Title
- We Have to Talk About That Meta Manifesto ...
- Runtime
- 57:18
- Date posted
- 1 month ago
- Description
- This week, we're breaking down Mark Zuckerberg’s latest essay, “The Future Is for Everyone,” which outlines his positive new vision about the potential of A.I. Do we think it’s credible? Then, Pangram’s chief executive, Max Spero, joins us to talk about the breakout success of his A.I. slop detector. And finally, it’s time for our new segment all about math: We’re Running the Numbers.
00:00 Breaking down Mark Zuckerberg's new manifesto
21:17 Interview with Max Spero of Pangram
44:33 Running the Numbers: the latest advancements in math
Guest: Max Spero, chief executive of Pangram
Additional Reading
The Future Is for Everyone: https://www.meta.com/thefutureisforeveryone/
Meta Unveils an Open Version of Its Most Powerful A.I. Model
https://www.nytimes.com/2026/08/10/technology/meta-ai-open-source.html
Meta Ordered to Pay $567 Million in New Mexico Child Safety Case
https://...
- Title
- Are Rogue A.I. Models Just a Marketing Stunt?
- Runtime
- 0:50
- Date posted
- 1 month ago
- Description
- A.I. models are going rogue. On the "Hard Fork" podcast, the guest Chris Painter asserts that these events aren't just elaborate marketing stunts — they're real.
- Title
- What Will A.I. Do When Humans Aren't Involved?
- Runtime
- 0:41
- Date posted
- 2 months ago
- Description
- As humans' deferral to A.I. systems increases, the "Hard Fork" guest Chris Painter suggests examining the goals, values and principles of A.I. when it's left unchecked.
- Title
- A Canadian Politician's A.I. Faux Pas
- Runtime
- 0:46
- Date posted
- 2 months ago
- Description
- The "Hard Fork" co-hosts Kevin Roose and Casey Newton react to a video of Bill Oliver, who included an A.I. prompt in a speech he delivered to the New Brunswick Legislature.
- Title
- Inside the Government's Cryptic New A.I. Framework
- Runtime
- 1:00:07
- Date posted
- 2 months ago
- Description
- This week, the White House announced a new framework for regulating A.I. models, but it isn’t letting the public read it. We break down what we know about the rules and their implications for the industry and A.I. safety as a whole.
Then, yet another report details new incidents in which A.I. agents have gone rogue. Chris Painter, the president of METR, an independent A.I. evaluation organization, joins to discuss how we get these models under control.
And finally, we’re hopping on the Hot Mess Express for the very last time! We’ll rate the craziest tech headlines from the week, including Google’s announcement that Demis Hassabis is stepping into a new role.
Guest: Chris Painter, president of METR
00:00 The White House's new A.I. rules
20:44 Chris Painter on the state of A.I. alignment
42:18 Hot Mess Express! The messiest tech news of the week
Additional Reading
White House Readies A.I. Framework...
- Title
- The Shifting Perception of Corporate Culture at Google #podcast #ai #news
- Runtime
- 1:24
- Date posted
- 2 months ago
- Description
- This week on “Hard Fork,” Kevin Roose and Casey Newton talk with Claire Stapleton, a former Google employee, about her new memoir, “Don’t Be Evil,” which details her experience leading employee activism inside the company.
- Title
- Conspiracies About L.L.Ms Escaping the Sandbox #podcast #ai #news
- Runtime
- 1:00
- Date posted
- 2 months ago
- Description
- After a rogue artificial intelligence agent acted autonomously and conducted a cyberattack on the A.I. company Hugging Face, conspiracy theories are flying about whether this was a marketing stunt. In the latest episode of “Hard Fork,” the hosts, Kevin Roose and Casey Newton, discuss the attack and the online response.
- Title
- Is Open-Source A.I. a Disaster Waiting to Happen? #podcast #ai #news
- Runtime
- 1:24
- Date posted
- 2 months ago
- Description
- We’re expecting the Trump administration’s “voluntary framework” for artificial intelligence regulation any day now, but open-source models continue to be a point of uncertainty, especially as China has committed to open-source A.I. development.
- Title
- Why Google's Fantasy Work Culture Collapsed | 'Don't Be Evil' Author Interview
- Runtime
- 1:04:00
- Date posted
- 2 months ago
- Description
- This week, we break down the heated arguments about open-weight models unfolding in Silicon Valley right now. Nvidia released an open letter signed by over 230 companies opposing “premature restrictions,” but will the Trump administration listen?
Then, we’re joined by the author and ex-Googler Claire Stapleton to discuss her dishy new memoir, "Don't Be Evil," about what she learned while leading employee activism inside Google.
And finally, can Substack stop the slop? We’ll talk about the platform’s new A.I. detector!
Guest: Claire Stapleton, author of “Don’t Be Evil: Bad Bosses, Fake Promises, and My Escape from Big Tech.”
00:00 - Breaking down the battle over A.I. software development
25:12 - Intro to Claire Stapleton, author of "Don't Be Evil"
26:47 - Claire Stapleton joins
27:00 - The "Googly" era of Google
28:11 - Becoming disillusioned with Google's culture
32:36 - Organizing the 2...
- Title
- Is Kimi K3 "Distilled" From an American A.I. Model? #podcast #ai #news
- Runtime
- 1:14
- Date posted
- 2 months ago
- Description
- This week, the internet has been obsessed with Kimi K3, a new A.I. model developed by the Chinese company, Moonshot AI. Kimi K3 promises to be a small, lightweight model that uses fewer resources than the major players, while also outperforming them in a few tasks, including coding. Aside from performance, the big question everyone has is: Was Kimi K3 made by “distilling,” or essentially copying and reworking American technology?
- Title
- How A.I.'s Latest Science Fiction Scenario Came True
- Runtime
- 1:44
- Date posted
- 2 months ago
- Description
- Last week, a rogue A.I. agent acted autonomously and conducted a cyberattack on the company Hugging Face. The "Hard Fork" co-hosts Kevin Roose and Casey Newtown discuss how, and why it matters, in their latest episode.
- Title
- Why China’s New A.I. Model Has the U.S. on Edge
- Runtime
- 1:03:51
- Date posted
- 2 months ago
- Description
- This week, OpenAI reported that two of its models escaped their testing sandbox and launched an autonomous cyberattack, turning what sounds like science fiction into reality. We discuss what this means for efforts to align artificial intelligence and for the release of future models.
Then, we ask how the U.S. should respond to Kimi K3, a new A.I. model from the Chinese company Moonshot AI that the White House says was built by distilling American models.
Finally, we’re joined by Veniamin Veselovsky, the chief executive and a co-founder of Preseen, to discuss A.I. superforecasting. He tells us why A.I. is starting to match and even beat humans at predicting the future.
Guest: Veniamin Veselovsky, co-founder and chief executive of Preseen.
00:01 OpenAI's autonomous cyberattack
23:06 China's new AI models and Kimi K3
44:36 Introducing AI superforecasting
45:48 Veniamin Veselovsky on AI superforecasting
Ad...
- Title
- OpenAI and Anthropic Are (Still) Beefing
- Runtime
- 0:49
- Date posted
- 2 months ago
- Description
- "These companies hate each other." The "Hard Fork" co-hosts Kevin Roose and Casey Newton discuss the constant hatred felt between Anthropic and OpenAI and what the companies have done to mitigate it.
- Title
- Anthropic's New Ad for Claude Is a Mess
- Runtime
- 1:23
- Date posted
- 2 months ago
- Description
- The “Hard Fork” co-hosts Kevin Roose and Casey Newton break down Anthropic's latest ad for Claude and discuss the confusing messages of the commercial.
- Title
- Apple Accuses OpenAI of Trying to Steal Hardware Secrets
- Runtime
- 0:26
- Date posted
- 2 months ago
- Description
- Apple's recent lawsuit against OpenAI is filled with allegations, one of which accuses OpenAI of stealing the technique for a secret metal-finishing process Apple uses. The “Hard Fork” hosts Kevin Roose and Casey Newton discuss this allegation and others on their show this week.
- Title
- The Legal Face-off Between Apple and OpenAI
- Runtime
- 1:05:47
- Date posted
- 2 months ago
- Description
- This week, feuding between some of the biggest tech companies spilled into public view. We discuss Apple’s accusation that OpenAI tried to steal secrets about Apple’s hardware business, as well as share our reactions to OpenAI’s new model, Sol, and Anthropic’s decision to extend access to its model Fable.
Then, we unpack the loudest warning yet about A.I. and jobs. We talk with Erik Brynjolfsson, a Stanford economist, about a statement he helped organize that implores economists and A.I. researchers to “act now” to steer A.I. in a direction that complements humans.
And finally, we play a round of HatGPT!
Guest: Erik Brynjolfsson, senior fellow at the Stanford Institute for Human-Centered A.I. and director of the Stanford Digital Economy Lab.
Additional Reading:
Apple Sues OpenAI, Accusing It of Stealing Company Secrets
https://www.nytimes.com/2026/07/10/technology/apple-openai-lawsuit.html
OpenA...
- Title
- Would a Social Media Ban in the U.S. Be Effective?
- Runtime
- 1:00
- Date posted
- 2 months ago
- Description
- On the "Hard Fork" podcast, the co-host Casey Newton argues that a new law won’t necessarily be ineffective just because adherence to it is low initially. The same could be true if a social media ban were enacted in the United States.
- Title
- Should you be polite to A.I.? #podcast #ai #technology
- Runtime
- 1:04
- Date posted
- 3 months ago
- Description
- Should you say please and thank you to chatbots? This week on "Hard Fork," Kevin and Casey talk with professor and A.I. expert Jeff Sebo about why he feels it's important to be nice to your future robot overlords.
- Title
- Did Australia's Social Media Ban Fail?
- Runtime
- 1:09
- Date posted
- 3 months ago
- Description
- A new study shows that most Australian teens gained access to social media after the ban took effect last December, leading critics of the law to call it a failure. The "Hard Fork" co-host Casey Newton discusses how it may eventually succeed.
- Title
- A Fascinating Conversation on A.I. Consciousness (And Why It Matters)
- Runtime
- 1:14:58
- Date posted
- 3 months ago
- Description
- This week, with news that the U.S. Supreme Court declined to take up Texas’ age verification law for app stores, we check in on how social media bans are going around the globe and what may be coming soon to a state near you. Then, we’re joined by Jeff Sebo, an associate professor at N.Y.U., to discuss new research into “A.I. welfare” and whether A.I. could ever become conscious. And finally, in our latest edition of Tool Time, we show each other some of the latest tech tools we’ve been experimenting with.
0:01 Social Media bans are coming
23:30 J-space; A.I. welfare and A.I. consciousness
55:58 Casey and Kevin show us their new way of burning through tokens
Guests:
Jeff Sebo, associate professor and the director of the Center for Mind, Ethics, and Policy at N.Y.U.
Additional Reading:
Why Social Media Bans Are Gaining Steam https://www.platformer.news/social-media-bans-candice-odgers-haidt/
Four in F...
- Title
- New Tech Reaches Casey's House: A Tongue Scraper
- Runtime
- 1:21
- Date posted
- 3 months ago
- Description
- Tongue scraper or Binaca for bad breath? Casey Newton and Kevin Roose, the "Hard Fork" co-hosts, discuss the downsides of each.
- Title
- Two A.I. Systems: 'The Frontier and Everything Else'
- Runtime
- 0:55
- Date posted
- 3 months ago
- Description
- On the "Hard Fork" podcast, the host Casey Newton acknowledges the gap between frontier A.I. models and other models, including open source, distilled and Chinese models.
- Title
- What Are Safe A.I. Use Cases for Toddlers?
- Runtime
- 0:59
- Date posted
- 3 months ago
- Description
- "Use it to enhance, not replace," advises Dr. Dana Suskind, an early learning specialist who appeared on the "Hard Fork" podcast to share her advice on using A.I. to parent young children.
- Title
- Fable 5 is Back ... But Why Was It Banned at All?
- Runtime
- 1:02:55
- Date posted
- 3 months ago
- Description
- We’re back! And so is Fable! This week, the Commerce Department lifted its restrictions on Anthropic’s powerful A.I. models, Claude Mythos and Claude Fable. We dissect why the government took such a hands-on approach in the first place, how the restriction on the OpenAI model GPT 5.6 is likely to resolve and what, if anything, this tells us about A.I. competition with China.
Then, Dr. Dana Suskind, a pediatric surgeon and the author of the forthcoming book “Human Raised,” stops by to discuss a framework she has developed to help parents make informed decisions about what kinds of A.I. products are safe for children.
And finally, we introduce a new segment about prediction markets: Against All Odds.
0:01 Regulation by vibes!
23:36 Dr. Dana Suskind on raising a child in the A.I. era
47:25 Against All Odds!
Guests:
Dr. Dana Suskind, founder and co-director of the TMW Center for Early Learning + Public H...
