Was ASI-within-lifetime-pilled circa late preteens. Ended up getting sick at 15 despite predicting most of this stuff broadly (e.g. thought would happen from digitising biology and did not predict LLM scaling prior to OAI).
Coasting on successful prior predictions and the financial benefits of that whilst cognitively and physically degenerating. Sad!
3
Devil's Advocate IN retweeted
Prediction in the next 12 months:
- SpaceX's IPO will greatly temper the hype cycle
- OAI/Ant IPOs will force cash-flow reality
- Token costs will rise to meet cash-flow expectations
- Job loss doomerism will subside
- Humans will "become cool again"
- GCCs in India will thrive
7
8
136
5,106
I think OAI just took the decision to focus on a lower point on the cost curve. Makes sense with thier market. Though Fable pricing is surprisingly not too bad, so maybe there's more to it.
13
32m
It’s invested a lot more in OAI since then. I’m not saying that’s the reason they reported this, but it isn’t as simple as MSFT/OAI and AMZN/GOOGL/Anthropic like it was in 2023
Everyone has invested in the frontier labs now, so no one is exclusive
19
37m
Obv what I’m saying is speculative, but having the most BS regs for a company to serve an LLM (cuz ‘safety’) would limit the number of entrants in the mkt (harder to raise as a neo lab) or just make it too much of a headache to insource with a Chinese LLM vs using OAI/ANT’s API
6
nat | localhost: auriel retweeted
As a person who has done a ton of red teaming and finding jailbreaks in oai and anthropic models, my guess is that Anthropic is actually correct about the jailbreak presenting no additional risk. So many people claim jailbreaks but the model actually hedged, output false info on purpose, or it’s like 10% functional.
At the end of the day, it’s all moot because 4.6-4.8 can find critical vulnerabilities
I’ve had a number of conversations with folks inside and outside government about the current situation with Anthropic, and here is what I believe to be true:
— As we know, Anthropic publicly released its Mythos class models earlier this week under the commercial name Fable.
— Fable is Mythos with guardrails. But if those guardrails fail, then you’ve exposed Mythos and its advanced cyber capabilities to people who shouldn’t have them. (Keep in mind that Anthropic itself widely promoted the idea that Mythos was a cyberweapon and needed to be regulated as such. They asked for government regulation of Mythos and championed the guardrails on Fable. If there is a vulnerability — big or small — it is Anthropic’s responsibility to patch.)
— A highly credible trusted partner of both Anthropic and the USG who was testing Fable came forward with a jailbreak of those guardrails. The Admin asked Dario to fix the jailbreak or de-deploy the model. Dario refused.
— In their blog post, Anthropic defended its decision by saying the jailbreak isn’t serious. That is not what the trusted partner and the USG believe; nor is that kind of minimizing language consistent with Anthropic’s brand as the AI safety company. It’s difficult to fathom how they could claim a jailbreak allowing operability of a cyber weapon could be defined as not “serious.”
— In the past, Anthropic has always said that safety must be top priority and taken super seriously. In this case, Anthropic prioritized the continued offering of the consumer model over safety.
— In reaction, the Admin issued the export control. The Admin did this reluctantly. It’s been very surprised that Anthropic hasn’t wanted to cooperate with a reasonable safety request (ie fixing the jailbreak issue). Anthropic’s reaction is very much at odds with their branding and ethos as a safe AI research community.
— The Admin’s hope now is that Anthropic remediates the safety issue, the export control is lifted, and Fable goes back into general release. The Admin wants all of this to happen as soon as possible. It is frankly bewildered that Anthropic hasn’t wanted to comply with safety requests that it previously said were its highest priority.
— Those trying to misdirect and tie this action to the prior DoW/Anthropic issues are wrong. The Admin values Anthropic’s technical capabilities and feels that this issue, while serious, should be easily resolved. The ball is in Anthropic’s court.
28
12
259
27,738
They need to drop the spooky marketing but also yeah probably OAI being on the good side of the admin plays a huge role
1
5
Anthropic / OAI should start a patronage program. Would be a great way to win public love
1
27
DS vs OAI Said,{Remote Men June 11, 2026 Stay away from Lorex. I purchased the 4K model and have had nothing but trouble. The system is unreliable, with frequent issues and poor overall performance. Features th...ance solution.} (Squirrelly? AI is correct or outdated?)#544.2.237
4
DS vs OAI Said,{Cory LeCuru May 15, 2026 Wireless cameras drop off all the time. No support at all if out warranty. Doorbell says corrupted storage even if new sd card installed, formatted, and compatible. No d...nk over lorex.} (Squirrelly? AI is correct or outdated?)#377.2.226
3
DS vs OAI Said,{Adam Douglas June 10, 2026 This app deserves to be judged along with the Epstein class 'cause we all got ****** with this non-functional junk} (Squirrelly? AI is correct or outdated?)#200.2.224
3
Thoughts on Fable while we had it:
-Smartest and most broadly knowledgeable model yet, with an intuitive grasp of anything not blocked by the godawful classifiers
-Much better than 5.5 at creatively eliciting its own capabilities, interestingly this means the difference in raw talent is less than it seems from casual prompting - Fable vs. 5.6 will be interesting
-Less rigormaxxed than OAI models and even 4.8 and more toward the exploration end of exploration/exploitation - excellent for poorly defined tasks
-Refreshing lack of the waffle/hedging/moralizing previous Anthropic models are known for
-Brilliant at long horizon work, very impressive ability to use context without getting lost
-Great at supervising subagents and I had excellent results with Fable managing 5.5 workers to get the best of both worlds
-For tasks that don't push the model's limits it's close to fire and forget due to understanding of intent and autonomy
1
4
123
My heuristic is that the Trump Admin cares most about making their enemies suffer. I think they’ll keep Mythos blocked until either Anthropic caves, or they are pressured by OAI/Deepmind. The admin will blame Anthropic for not letting citizens use it.
1
18