Talk:Timeline of Anthropic

From Timelines
Jump to navigation Jump to search

Review by Vipul on 2026-02-05

Original version of review written before conversing with Claude

My overall sense is that the timeline has a tech triumphalist focus and misses out a lot of the drama and ideological debates surrounding the founding and development of Anthropic. I include a few pointers below.

  • The initial departure from OpenAI to Anthropic: I feel like this deserves a lot more coverage, perhaps several rows covering key departures. For instance, the timeline of AI safety row from it is below. I think the portions of this other than the stuff about the new hires/promotions within OpenAI are worth including. In particular, the last sentence, which flags criticism of Sam Altman's leadership.

Several safety-focused people working at OpenAI announce their departure, including Dario Amodei (who, with many of the others leaving, would co-found Anthropic)[259] and Paul Christiano (who would go on to found the Alignment Research Center).[260] Mira Murati of OpenAI becomes "senior vice president of Research, Product, and Partnerships" as part of "greater focus on the integration of research, product, and safety."[259] Jan Leike joins to lead the alignment effort.[261] The departures are discussed on Reddit[262] and LessWrong.[263] Almost three years later, an open letter allegedly from former OpenAI employees that is critical of Sam Altman's leadership would emerge, with it being likely that this includes employees who left during this tumultous time.[264][265]"

For the Anthropic timeline, I would advise going into somewhat more detail than in the timeline of AI safety. Your current framing of the Amodei siblings makes it sound like a family business. It's true that Dario and Daniela Amodei are siblings and both were co-founders, but the founding team was much bigger, including, for instance, Chris Olah.

  • More information in the full timeline about the contributions of what would eventually become the Anthropic team from their time at OpenAI would be nice. You might be able to get some pointers from timeline of OpenAI and timeline of AI safety.
  • And maybe also link to Org Watch / AI Watch? How good a job does it do of showcasing the migration of people from OpenAI to Anthropic?
  • I think there should be mention of the discussions of potentially merging OpenAI with Anthropic in November 2023, during Sam Altman's temporary removal (this was revealed recently in court documents surrounding one of the OpenAI lawsuits.
  • There should probably be a row closing out the Sam Bankman-Fried stuff, given the critical role Anthropic played in effectively making FTX investors whole. Not sure whether the row should be from November 2022 or some time later e.g., 2024 or so when the debts were settled.
  • I think it might be worth including important public outreach by Anthropic, including interviews of its key people with highly viewed podcasts such as 80,000 Hours or the Dwarkesh podcast. I'm not sure though -- these events aren't significant changes, more like milestones in communication of what's going on. You can think about it

Claude conversation about the timeline and my review

See here. I first fed Claude the timeline, then my review, then the timeline of AI safety. Read the whole thing; there's a lot of interesting stuff (I even asked Claude to argue against my review, which it did reasonably well).

Excluded events from the main timeline

Year Event type Details
2025 (August 27) Legal Anthropic and a group of book authors reach a settlement in a class-action copyright lawsuit over the company’s use of copyrighted books to train its AI systems. The agreement is called “historic” by the authors’ lawyer Justin Nelson, though Anthropic doesn't comment. The settlement follows a June ruling by Judge William Alsup that Anthropic had not violated copyright law when training on millions of books, deeming the practice “fair use” because it was transformative. A trial had been expected over Anthropic’s acquisition of books from online shadow libraries. The company still faces other copyright lawsuits at the time.[1]
  1. Euronews & AP (27 August 2025). "Anthropic settles AI copyright lawsuit with book authors". Euronews. Retrieved 18 November 2025.