Log inSign up
Jason Wolfe
1,391 posts
user avatar
Jason Wolfe
@w01fe
alignment and the model spec @OpenAI (opinions are my own)
Joined May 2010
759
Following
3,593
Followers
  • user avatar
    Jason Wolfe
    @w01fe
    Feb 12, 2025
    Excited that our updated Model Spec is out, and now open source! We've worked hard to make the principles clearer and more detailed, and maximize intellectual freedom for users and developers while ensuring our models stay within bounds. Please send your feedback!
    user avatar
    OpenAI
    @OpenAI
    Feb 12, 2025
    Today we're sharing a major update to the Model Spec—a document which defines how we want our models to behave. The update reinforces our commitments to customizability, transparency, and intellectual freedom to explore, debate, and create with AI. openai.com/index/sharing-…
    39K
  • user avatar
    Jason Wolfe
    @w01fe
    Feb 13, 2025
    Replying to @TheZvi
    😊 believe it or not, this is an error that was introduced while copy editing. Thanks for pointing it out, will aim to fix in the next version!
    2.7K
  • user avatar
    Jason Wolfe
    @w01fe
    Jul 24, 2015
    I made a thing. And then I used it to delete 10% of our Clojure codebase.
    GitHub - w01fe/sniper: emacs assistant for sniping dead code.
    From github.com
  • user avatar
    Jason Wolfe
    @w01fe
    Apr 11, 2025
    We just released a small bugfix update to the OpenAI Model Spec. Key changes: - Fixed a copy-editing error where "misalignment" wasn't listed as a reason assistants might pursue incorrect goals. - Clarified that "Do not lie" allows "pleasantries" rather than "white lies." 1/2
    4.3K
  • user avatar
    Jason Wolfe
    @w01fe
    Nov 19, 2023
    ❤️
    user avatar
    Sam Altman
    OpenAI
    @sama
    Nov 19, 2023
    i love the openai team so much
    5K
  • user avatar
    Jason Wolfe
    @w01fe
    Sep 16, 2013
    Google's first open-source Clojure project? github.com/google/hesokuri
  • user avatar
    Jason Wolfe
    @w01fe
    Dec 24, 2015
    Hi friends: I'm looking for remote consulting gigs, preferably w/ML and/or Clojure. If you (or a friend) wants to chat, say hi@w01fe dot com
  • user avatar
    Jason Wolfe
    @w01fe
    Apr 11, 2025
    Replying to @w01fe
    Model Spec: model-spec.openai.com/2025-04-11.html Changelog: github.com/openai/model_s… Special thanks to @TheZvi for bringing the above issues to our attention!
    Abstract art featuring a striking contrast of teal and pink brushstrokes on an orange canvas, overlaid with the white OpenAI logo.
    OpenAI Model Spec
    From model-spec.openai.com
    1K
  • user avatar
    Jason Wolfe
    @w01fe
    Dec 2, 2016
    Had a great first day @SemMachines! Very excited to work with this incredible team to help build the next gen of conversational interfaces.
  • user avatar
    Jason Wolfe
    @w01fe
    Feb 14, 2025
    Replying to @TheZvi
    The intention was "The assistant might pursue the wrong objective due to misalignment, misunderstanding ...". When "Misalignment" was pulled up into a list header for clarity, it was dropped from the list of potential causes, unintentionally changing the meaning.
    385
  • user avatar
    Jason Wolfe
    @w01fe
    Mar 6, 2023
    I am thrilled to announce that I'll be joining @OpenAI this week! So excited to learn from the team and contribute to such an ambitious mission.
    1.3K
  • user avatar
    Jason Wolfe
    @w01fe
    Sep 21, 2020
    Really excited to finally share details about the conversational AI approach that we've developed at Microsoft Semantic Machines (including a TACL paper, dataset, and leaderboard). In addition to the product and NLP sides, I've found the PL bits lots of fun.
    user avatar
    Microsoft Research
    Microsoft
    @MSFTResearch
    Sep 21, 2020
    Researchers at Microsoft Semantic Machines are taking a new approach to conversational AI—modeling dialogues with compositional dataflow graphs. Learn how the framework supports flexible, open-ended conversations, and explore the dataset and leaderboard: aka.ms/AA9oxf3
  • user avatar
    Jason Wolfe
    @w01fe
    Feb 1, 2013
    The video for my Strange Loop 2012 talk is up: "Graph: Composable Production Systems in Clojure" bit.ly/UI4eL1 #prismatic #clojure
  • user avatar
    Jason Wolfe
    @w01fe
    Feb 12, 2025
    Replying to @w01fe
    It was super fun to work on this with the amazing @Laurentia___ , @joannejang , @boazbaraktcs, Rodrigo Perez, and Andrea Vallone, and so many others at OpenAI.
    572

New to X?

Sign up now to get your own personalized timeline!

Create account

By signing up, you agree to the Terms of Service and Privacy Policy, including Cookie Use.

Terms·Privacy·Cookies·Accessibility·Ads Info·© 2026 X Corp.
Don't miss what's happening
People on X are the first to know.
Log inSign up