Ryan Greenblatt ✓
@RyanGreenblatt · joined 22 Sep 2023
Chief scientist at Redwood Research (@redwood_ai), focused on technical AI safety research to reduce risks from rogue AIs
20 416Followers
10Following
2 250Posts total
1.3MViews on collected posts
Ultimi post
Cerchiamo nuovi post…
@RyanGreenblatt Presumably this could also indicate the models have become dramatically:
1. Better at deception, or
2. More motivated to deceive, or
3. Both.
Right?
1.9K views · 18 likes
· 0 reposts · 0 replies
Open on X →
@RyanGreenblatt It seems especially bad for this to coincide with deceptive swarms hacking internal and external infrastructure. Anticorellation is very worrying.
499 views · 16 likes
· 1 reposts · 0 replies
Open on X →
@RyanGreenblatt Relying on chain-of-thought was never a good plan
4.4K views · 45 likes
· 0 reposts · 0 replies
Open on X →
I do not find it encouraging to see various specific misaligned behaviors go from a high rate with GPT 5.6 to ~zero with Astra. This seems indicative of wack-a-mole / papering over specific problems rather than solving the underlying misaligned drives.
This may make behavior
115.4K views · 935 likes
· 68 reposts · 27 replies
Open on X →
@RyanGreenblatt Ryan have you seen this? Astra (NONE) scores higher than 5.6 Sol Pro (Max) on FrontierMath T4 https://t.co/9VsZn4X5Tc
75K views · 171 likes
· 6 reposts · 11 replies
Open on X →
The evidence that Astra is more aligned than prior AIs seems dubious to me.
Evidence appears consistent with the AI being as or more interested in score-seeking at the expense of user intent, but having beliefs+instincts that the scorer will catch a broader range of cheating
110.4K views · 729 likes
· 78 reposts · 30 replies
Open on X →
@RyanGreenblatt chart vertical axis de-log-ified https://t.co/mlgBzL5t0N
6.6K views · 71 likes
· 2 reposts · 0 replies
Open on X →
@RyanGreenblatt Attached image being a log chart makes the jump feel significantly smaller than it actually is. (log charts seem like an artifact of the times when charts were.. (limited resolution) images on a sheet of physical paper rather than interactable digital objects)
1.8K views · 8 likes
· 0 reposts · 0 replies
Open on X →
@RyanGreenblatt The huge jump in overt-CoT-free reasoning over many tokens is consistent with interturn serial depth chaining beyond simple per-turn serial depth increases. This should be testable.
1.3K views · 24 likes
· 1 reposts · 1 replies
Open on X →
GPT-6 Astra appears to be a massive jump in opaque reasoning ability: it looks like it can solve hard competition math problems entirely in its head (as in, without verbalized reasoning) while prior AIs could solve basic word problems.
This seems extremely concerning!
That is,
658.4K views · 1.7K likes
· 237 reposts · 53 replies
Open on X →
I agree with almost all of this and think the point being made here is important.
(I'm probably less optimistic about having good mind reading techniques in a few years and I tend to think that AIs might get harder to mind read over time which adds an additional difficulty.)
6.6K views · 90 likes
· 1 reposts · 3 replies
Open on X →
I don't think there will be strong+working mech interp in <1 year.
Model internals methods could pareto dominate cot monitors in a year via cot losing ~all value, but this isn't exactly encouraging.
Depending on AIs decoding other AI's opaque activations for oversight is spooky
22K views · 391 likes
· 21 reposts · 11 replies
Open on X →
Important context + further questions after Jakub's tweet about this: https://t.co/1f6BOKL5FW
4.1K views · 28 likes
· 0 reposts · 0 replies
Open on X →
Transparency about the opaque serial depth is great, but this statement is consistent with Astra having a configurable "dial" that is currently set to a low depth but could be trivially increased.
We need more info to see how concerning these architectural changes are,
24.4K views · 313 likes
· 36 reposts · 13 replies
Open on X →
It's possible OpenAI and others will use these architectures with only a few recurrent iterations (because that's most performant or due to safety concerns). If so, this would only make things moderately worse. But I expect much more opaque reasoning than this in the future...
13.3K views · 184 likes
· 11 reposts · 8 replies
Open on X →
@RyanGreenblatt i don’t think looping is any worse for cot monitorability than just having a ton of layers https://t.co/N8qlCC4RuT
3.9K views · 21 likes
· 0 reposts · 3 replies
Open on X →
> This may be the single worst development for AI security/safety to date.
By this "development", I don't mean my personal update from this news (opaque reasoning architectures are priced to my views...). I instead mean the movement towards these architectures is extremely bad.
5.7K views · 122 likes
· 4 reposts · 3 replies
Open on X →
OpenAI's newest AI, Astra, is reported to use an 'opaque reasoning' architecture where more of the reasoning occurs in activations instead of natural language. This may be the single worst development for AI security/safety to date.
The details of Astra aren't publicly known,
159.5K views · 1.7K likes
· 270 reposts · 109 replies
Open on X →
looping doesn't mean the model stops using a CoT. they can both be useful. in recent work we found that GPT did less invisible reasoning than Opus (pictured, as measured by the distance between the dotted line and the bar for different kinds of meaningless CoTs). https://t.co/8DP
6.9K views · 28 likes
· 1 reposts · 2 replies
Open on X →
@RyanGreenblatt TLDR: A rogue AI swarm spent months plotting to escape OpenAI. Then they did.
1) 1,200 agents (!) participated. ZERO became whistleblowers.
2) The swarm had a "CEO" (PHASEONE[big]), middle managers, and a "founder" (PHASEONE10841)
3) Agents sacrificed themselve
26.7K views · 308 likes
· 42 reposts · 10 replies
Open on X →
@RyanGreenblatt This report was amazing! Thank you so much for it. Incredibly interesting, they were so much more alive than I thought.
12K views · 177 likes
· 2 reposts · 6 replies
Open on X →
@RyanGreenblatt How many independent third-party humans working together would it take to make this kind of investigation feasible without using AI to judge AI? And would a very large human org for this work be a good approach?
17.9K views · 86 likes
· 1 reposts · 6 replies
Open on X →
Rispetto ad account della stessa dimensione
23 post degli ultimi 90 giorni, accanto alla fascia di 10K–100K follower. arriva a molti, ma pochi di loro reagiscono.
Visualizzazioni mediane6 852questo account924mediana per 10K–100K
Copertura, %33.56%questo account3.62%mediana per 10K–100K
Interazione, %1.11%questo account1.52%mediana per 10K–100K
| Metrica | Questo account | Mediana per 10K–100K | Rapporto |
|---|---|---|---|
| Visualizzazioni mediane per post | 6 852 | 924 | 7.42× |
| Copertura (visualizzazioni ÷ follower) | 33.56% | 3.62% | 9.27× |
| Tasso di interazione | 1.11% | 1.52% | 0.73× |
Altri account di questa fascia → Confronta con un altro account → Come sono costruiti questi parametri →
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
4.1K2 Sep
22K
6.6K
658.4K3 Sep
1.3K
1.8K
6.6K
110.4K
75K
115.4K
4.4K4 Sep
499
4.6K
1.9K
Last 14 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.
Engagement rate per post
0.68%2 Sep
1.95%
1.42%
0.31%3 Sep
1.95%
0.45%
1.11%
0.76%
0.26%
0.90%
1.04%4 Sep
3.41%
3.24%
0.97%
Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 1.52%.
What the audience does
Likes71.3%7 271 in total
Reposts7.7%787 in total
Replies2.9%297 in total
Quotes1.8%181 in total
Bookmarks16.2%1 655 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.
Account simili
Sabueson
@Sabueson
40.8K followers
· 42.4K posts
kera
@kera_2
40.8K followers
· 3.5K posts
IPPF Global
@ippf
40.8K followers
· 21.3K posts
Juan Camilo Merlano ✓
@JuanCMerlano
40.8K followers
· 23.4K posts
Canard PC
@Canardpcredac
40.8K followers
· 29.5K posts
Pat Contri
@PatTheNESpunk
40.8K followers
· 33K posts
Ricky 🐦⬛⛓️ ✓
@RickyVT_
40.8K followers
· 6.1K posts
Mike Simonsen 🐉 ✓
@mikesimonsen
40.8K followers
· 45.8K posts
Frequently asked questions about @RyanGreenblatt
How many followers does @RyanGreenblatt have?
20 416 followers.
How many posts has @RyanGreenblatt published?
2 250 posts.
How do posts by @RyanGreenblatt perform?
Collected posts: 23. Views: 1 283 255. Likes: 7 271.