Human and robot coach. Follow me at https://xitter.cf/t.co/rgbs4wAtqD

Joined July 2013
New: A deep dive in what's wrong with the current framework of intelligence, power and alignment in AI research. substack.com/home/post/p-151…
2
174
thecrowdview retweeted
It's quite striking that despite everything that's happened in AI over the last 3 years, the world is still spending *less* capex building semiconductor foundries today than in 2022 All of AI is still small enough to be washed away by consumers buying -10% fewer Android phones
10
14
3
201
21,539
It is important to note what generations of masters know that modern folks forget: consciousness is never a tool as it is not within your control.
I deleted the earlier tweet on whistleblowing as it was being pulled out of context. TBC: This isn't a new Claude feature and it's not possible in normal usage. It shows up in testing environments where we give it unusually free access to tools and very unusual instructions.
1
33
I’m really fed up by the narrative that in the age of AI, there will be no more human work to be had. Intelligence consumes problems, and therefore when intelligence is abundant, great problems will be scarce. I have yet to meet an AI that has problem finding skills.
1
1
45
There are huge demands even just for problems in testing AI, most current benchmarks suck. Not to mention to understand what are the problems that will be valuable to the world.
1
31
Of course, I understand that there will not be work for everyone. But look around, that already happened for a while, with or without the new AI era.
18
thecrowdview retweeted
Fail fast might be the wrong thing now that the idea maze has shifted. Just use interesting technology, follow your own curiosity, and if you live at the edge of the future like PG says, there's so many great startup ideas you're very likely to just bump into one.
55
121
21
1,415
143,266
thecrowdview retweeted
Cursor writes almost 1 billion lines of accepted code a day. To put it in perspective, the entire world produces just a few billion lines a day.
225
282
167
5,107
771,302
thecrowdview retweeted
Today we are announcing ARC-AGI-2, an unsaturated frontier AGI benchmark that challenges AI reasoning systems (same relative ease for humans). Grand Prize: 85%, ~$0.42/task efficiency Current Performance: * Base LLMs: 0% * Reasoning Systems: <4%
66
317
87
2,329
462,602
Nail art is getting so crazy these days
48
104
13
806
56,819
thecrowdview retweeted
Anyone paying attention to AI on X is likely year(s) ahead of the rest of the world. I regularly show demos that are fully known by everyone here to more offline people and look like a time traveler. The next few years will be wild.
When I spend time with people outside of tech, I am convinced AI is UNDERhyped Show DeepResearch to a Professor Show an AI Nurse to an understaffed hospital administrator An AI reservations agent to an understaffed call center Most people have NO IDEA…and are blown away
199
135
25
1,781
196,333
thecrowdview retweeted
"The quality of your life is determined by the quality of your thoughts." – Marcus Aurelius
962
4,058
242
30,678
1,418,918
thecrowdview retweeted
gg there are only 7 American coders better than o3 (benq, ecnerwala, a_g, neal, conqueror_of_tourist, crystally, rqi)
41
36
31
539
183,967
The final admission that the 2023 strategy of OpenAI, Anthropic, etc. (“simply scaling up model size, data, compute, and dollars spent will get us to AGI/ASI”) is no longer working!
OPENAI ROADMAP UPDATE FOR GPT-4.5 and GPT-5: We want to do a better job of sharing our intended roadmap, and a much better job simplifying our product offerings. We want AI to “just work” for you; we realize how complicated our model and product offerings have gotten. We hate the model picker as much as you do and want to return to magic unified intelligence. We will next ship GPT-4.5, the model we called Orion internally, as our last non-chain-of-thought model. After that, a top goal for us is to unify o-series models and GPT-series models by creating systems that can use all our tools, know when to think for a long time or not, and generally be useful for a very wide range of tasks. In both ChatGPT and our API, we will release GPT-5 as a system that integrates a lot of our technology, including o3. We will no longer ship o3 as a standalone model. The free tier of ChatGPT will get unlimited chat access to GPT-5 at the standard intelligence setting (!!), subject to abuse thresholds. Plus subscribers will be able to run GPT-5 at a higher level of intelligence, and Pro subscribers will be able to run GPT-5 at an even higher level of intelligence. These models will incorporate voice, canvas, search, deep research, and more.
53
109
38
1,282
310,914
thecrowdview retweeted
When I wrote about Humanity’s Last Exam, the leading AI model got an 8.3%. 5 models now surpass that, and the best model gets a 26.6%. That was 10 DAYS AGO.
It looks like the latest OpenAI model is very doing well across many topics. My guess is that Deep Research particularly helps with subjects including medicine, classics, and law.
86
149
32
1,060
262,323
This hasn’t aged well.
CEO of OpenAi, Sam Altman: “It’s hopeless to compete with us”
67
i know a waymo hate to see this one coming
10
37
39
674
646,407
thecrowdview retweeted
This is what would-be founders should be doing when they're young, not bogus startup pitching competitions.
This Project got me an interview at Google
130
448
39
8,169
697,492