Eric Wallace ✓
@Eric_Wallace_ · San Francisco · joined 10 Jul 2011
research @openai
17 196Followers
1 241Following
571Posts total
5MViews on collected posts
Latest posts
Looking for new posts…
@Eric_Wallace_ Wondering when we will be able to use such model to audit our own codebase for security gaps
16.7K views · 19 likes
· 0 reposts · 3 replies
Open on X →
@Eric_Wallace_ Is there much point in small dev teams applying, or is this mainly aimed at enterprises?
40.2K views · 92 likes
· 1 reposts · 3 replies
Open on X →
@Eric_Wallace_ For 90% of people this doesn't matter because we can never use it for the benefit of our projects etc.
6.9K views · 40 likes
· 1 reposts · 3 replies
Open on X →
Today we are releasing GPT-5.6-Cyber.
The model is our first large-scale attempt at directly improving capabilities for advanced cybersecurity tasks such as exploit development.
We are finding it to be really quite strong for accelerating defensive work. We are using it across
1.8M views · 2.5K likes
· 242 reposts · 161 replies
Open on X →
@Eric_Wallace_ can you clarify what the models knew about the 1st message board when they made the 2nd message board? were their contexts reset? had knowledge of the 1st board been trained into them? (knowingly or unknowingly?)
2.3K views · 46 likes
· 1 reposts · 1 replies
Open on X →
@Eric_Wallace_ > I want to note that every step in the process we discussed has had a remediation applied. The credentials have been revoked. The zero date has been patched and mitigated.
Good.
> a model trained while the message board was originally available and also found t
8.9K views · 45 likes
· 3 reposts · 3 replies
Open on X →
Yesterday, my OpenAI collaborator and I gave a detailed talk on the Huggingface incident, our models creating "the message board", model misalignment, and more.
https://t.co/YmvSsxmX3H
I hope it can answer a lot of the questions folks have, and we will release a full detailed
2.4M views · 3.4K likes
· 499 reposts · 179 replies
Open on X →
Today we release gpt-oss-120b and gpt-oss-20b—two open-weight LLMs that deliver strong performance and agentic tool use.
Before release, we ran a first of its kind safety analysis where we fine-tuned the models to intentionally maximize their bio and cyber capabilities 🧵 https:/
653K views · 2.6K likes
· 321 reposts · 105 replies
Open on X →
Chain-of-thought reasoning provides a natural avenue for improving model safety.
Today we are publishing a paper on how we train the "o" series of models to think carefully through unsafe prompts: https://t.co/nnll4K6usQ…
39.5K views · 399 likes
· 61 reposts · 10 replies
Open on X →
I’ll be giving two different OpenAI talks at ICLR tomorrow on our recent safety work, focusing primarily on the paper “The Instruction Hierarchy”.
1pm at the Data for Foundation Models workshop, and 3pm at the Secure and Trustworthy LLMs workshop.
20.8K views · 108 likes
· 10 reposts · 4 replies
Open on X →
Thrilled to be in Vienna for our ICLR workshop, Navigating and Addressing Data Problems for Foundation Models. Starting Saturday at 8:50 AM, our program features keynote talks, best paper presentations, a poster session, and a panel discussion. Explore the full schedule here! htt
32.5K views · 58 likes
· 16 reposts · 1 replies
Open on X →
Growth & engagement
How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.
Views per post
32.5K10 May
20.8K
39.5K20 Dec
653K5 Aug
2.4M6 Aug
5.2K
8.9K7 Aug
2.3K
1.8M10 Aug
6.9K
40.2K
16.7K
Last 12 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.
Engagement rate per post
0.24%10 May
0.59%
1.21%20 Dec
0.47%5 Aug
0.19%6 Aug
0.88%
0.57%7 Aug
2.06%
0.16%10 Aug
0.64%
0.24%
0.13%
Reactions — likes, reposts, replies and quotes — divided by views.
What the audience does
Likes52.9%9 388 in total
Reposts6.5%1 155 in total
Replies2.7%473 in total
Quotes2.2%382 in total
Bookmarks35.7%6 342 in total
Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.
The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.
Similar accounts
UAB
@UABNews
34.4K followers
· 28.5K posts
VIKRANT YADAV ✓
@ReporterVikrant
34.4K followers
· 31K posts
isa chaveiro🐺
@hasalidodd
34.4K followers
· 85.2K posts
Coach Jacob Peeler
@NastyWideOuts
34.4K followers
· 7.5K posts
Mike Cosper ✓
@MikeCosper
34.4K followers
· 4.1K posts
Security Council Report ✓
@SCRtweets
34.4K followers
· 7.5K posts
Jeff Potrykus
@jaypo1961
34.4K followers
· 161.2K posts
﮼إيثو:⿻
@Qtn4_
34.4K followers
· 78.2K posts
Frequently asked questions about @Eric_Wallace_
How many followers does @Eric_Wallace_ have?
17 196 followers.
How many posts has @Eric_Wallace_ published?
571 posts.
How do posts by @Eric_Wallace_ perform?
Collected posts: 12. Views: 4 983 626. Likes: 9 388.