Hard Fork
The New York Times · Technology
YAPP has Smart‑Skip data for 38 episodes of Hard Fork. Recent episodes average 4.8 minutes of ads per 69.7‑minute episode.
Listen to the good stuff in Hard Fork
YAPP maps every episode before you press play, so you land on the parts you came for. Listen to the good stuff, skip the rest. 10 free listens, then a 60-day trial.
Recent episodes
| Episode | Published | Length | Ad time | Breaks |
|---|---|---|---|---|
| A.I. Safety Goes Mainstream + a ‘Hard Fork’ Exit AMA | 1:30:01 | 7:18 | 6 | |
| The Ezra Klein Show: The A.I. Revolt Is Here | 1:24:06 | 5:07 | 5 | |
| The A.I. Mob That Attacked Hugging Face + METR’s Ajeya Cotra | 1:23:37 | 5:15 | 6 | |
| Meta Shifts the Blame + Do Data Center Bans Work? + The Final HatGPT | 1:08:14 | 5:26 | 5 | |
| OpenAI’s Two-Week Pause + Jill Lepore on the Threat of the “Artificial State” + Train of Thought | 1:07:44 | 4:36 | 4 | |
| Zuckerberg’s Anti-Doom Fantasy + Finally an A.I. Detector That Works + A.I. Math | 1:08:20 | 5:09 | 5 | |
| The White House’s Secret A.I. Rules + The State of Model Alignment With METR’s Chris Painter + The Final Hot Mess Express | 1:10:57 | 5:38 | 6 | |
| Open Model Wars + Claire Stapleton's Dishy Google Memoir + Substack's Slop Fight | 1:11:32 | 5:35 | 5 | |
| OpenAI Models Go Rogue + Kimi K3 Freakout + A.I. Superforecasting | 1:08:14 | 1:23 | 1 | |
| Do Social Media Bans Work? + A Conversation About A.I. Consciousness + Tool Time | 1:23:47 | 5:27 | 6 | |
| Fable Ban Reversed + Dr. Dana Suskind on Parenting With A.I. + Prediction Market Drama | 1:11:01 | 4:52 | 5 | |
| ‘The Daily’ and ‘The Opinions’: How A.I. Is Changing Loneliness and Taste | 1:05:30 | 7:53 | 7 | |
| ‘Hard Fork’ Live, Part 3: Differing Visions of an A.I. Future | 59:55 | 4:55 | 5 | |
| ‘Hard Fork’ Live Part 2: Dylan Field on Standing Out in the A.I. Era | 35:29 | 5:43 | 5 | |
| ‘Hard Fork’ Live, Part 1: Satya Nadella and Cindy Cohn | 1:09:01 | 3:07 | 4 | |
| Hot I.P.O Summer + What Is A.I. Doing to Math? + HatGPT | 1:09:00 | 4:51 | 7 | |
| Interesting Times: Why Are We Still Driving? | 1:03:51 | 6:06 | 5 | |
| Our Field Trip to Google I/O + A Sit-Down With Sundar Pichai + System Update | 58:38 | 5:19 | 5 | |
| A.I. Safety Is So Back + Mythos Mayhem with Nikesh Arora + Hot Mess Express | 1:10:19 | 4:22 | 4 | |
| Can the U.S. Rein in Prediction Markets? + Joanna Stern on Her Year of A.I. Experiments + Our Producer Goes to Attention School | 1:15:36 | 3:44 | 5 | |
| OpenAI’s Big Reset + A.I. in the Doctor’s Office + Talkie, a pre-1930s LLM | 1:11:28 | 1:18 | 3 | |
| Tim Cook’s Legacy + The Future of U.B.I. With Andrew Yang + HatGPT | 1:18:10 | 4:00 | 4 | |
| A.I. Backlash Turns Violent + Kara Swisher on Healthmaxxing + The Zuck Bot Is Coming | 1:05:01 | 3:32 | 3 | |
| Anthropic’s Cybersecurity Shock Wave + Ronan Farrow and Andrew Marantz on Their Sam Altman Investigation + One Good Thing | 1:08:26 | 5:56 | 5 | |
| The Future of Addictive Design + Going Deep at DeepMind + HatGPT | 1:15:29 | 6:41 | 6 |
Ad load varies by listener. Most podcast advertising is dynamically inserted, so figures reflect ad load as observed by YAPP's processing pipeline at the time each episode was fetched, not what any individual listener hears. Methodology
Listening to Hard Fork with YAPP
Hard Fork runs 4.9 ad breaks in a typical episode, so there is a fair amount between you and what you came for. Across the episodes we have processed, that comes to about 4.8 minutes per episode, or 4.2 minutes for every hour of listening.
YAPP maps each episode before you play it, so you get the show and not the rest of it. Because the ads are stitched in per listener, YAPP matches what you are actually hearing rather than replaying a fixed set of timestamps.
How these numbers are measured
YAPP transcribes each episode, classifies advertising segments, and refines their boundaries using audio fingerprinting. Figures cover the 38 episodes of this show that YAPP has processed. Because most podcast ads are dynamically inserted and vary by listener, these figures reflect what our pipeline observed rather than what every listener hears. Full methodology
Boundaries for 16% of detected segments were refined by audio fingerprinting; the rest rest on transcript timing and are less precise.
Not seeing your show?
Tell us which one and we will let you know when it is ready.
Are you the creator of this show?
You can claim it, send us a correction, or opt out entirely. Creator options