o3 mightily beats Gemini, Opus 4 and others in the game of Diplomacy
only Gemini was able to win even one game, due to o3’s “ruthless” strategies
o3: ruthless
Gemini Pro: alliance builder
Opus 4: overly honest
www.reddit.com/r/singularit...
only Gemini was able to win even one game, due to o3’s “ruthless” strategies
o3: ruthless
Gemini Pro: alliance builder
Opus 4: overly honest
www.reddit.com/r/singularit...
From the singularity community on Reddit: o3 is the top AI Diplomacy player, followed by Gemini 2.5 Pro
Explore this post and more from the singularity community
www.reddit.com
June 7, 2025 at 11:41 AM
o3 mightily beats Gemini, Opus 4 and others in the game of Diplomacy
only Gemini was able to win even one game, due to o3’s “ruthless” strategies
o3: ruthless
Gemini Pro: alliance builder
Opus 4: overly honest
www.reddit.com/r/singularit...
only Gemini was able to win even one game, due to o3’s “ruthless” strategies
o3: ruthless
Gemini Pro: alliance builder
Opus 4: overly honest
www.reddit.com/r/singularit...
I keep getting asked about my take on these CO2 estimates for the o3 model by the press and members of the community, so I'll interrupt my vacation to comment 🤓
TL;DR- any kind of estimate is a proxy, and instead of wasting our time and energy, we should demand👏🏼 accountability👏🏼
TL;DR- any kind of estimate is a proxy, and instead of wasting our time and energy, we should demand👏🏼 accountability👏🏼
December 29, 2024 at 9:33 AM
I keep getting asked about my take on these CO2 estimates for the o3 model by the press and members of the community, so I'll interrupt my vacation to comment 🤓
TL;DR- any kind of estimate is a proxy, and instead of wasting our time and energy, we should demand👏🏼 accountability👏🏼
TL;DR- any kind of estimate is a proxy, and instead of wasting our time and energy, we should demand👏🏼 accountability👏🏼
One thing we’re losing with LLMs is a plausible reason to ask the hive mind questions just slightly too obscure for a search engine. “Blegs” used to be great for community on social media. But now I take them to o3.
June 3, 2025 at 1:25 PM
One thing we’re losing with LLMs is a plausible reason to ask the hive mind questions just slightly too obscure for a search engine. “Blegs” used to be great for community on social media. But now I take them to o3.
"When trying to develop a measure of intelligence, it’s essential to avoid Goodhart’s law: “When a measure becomes a target, it ceases to be a good measure.” As a community of AI researchers, we really need to figure that one out." @melaniemitchell.bsky.social
aiguide.substack.com/p/did-openai...
aiguide.substack.com/p/did-openai...
Did OpenAI Just Solve Abstract Reasoning?
OpenAI’s o3 model aces the "Abstraction and Reasoning Corpus" — but what does it mean?
aiguide.substack.com
December 23, 2024 at 9:32 AM
"When trying to develop a measure of intelligence, it’s essential to avoid Goodhart’s law: “When a measure becomes a target, it ceases to be a good measure.” As a community of AI researchers, we really need to figure that one out." @melaniemitchell.bsky.social
aiguide.substack.com/p/did-openai...
aiguide.substack.com/p/did-openai...
Review of methane through space and time, coauthored with @planetdr.bsky.social, is out (OA) in @annualreviews.bsky.social.
Hopefully a helpful resource for community and courses. www.annualreviews.org/content/jour...
Hopefully a helpful resource for community and courses. www.annualreviews.org/content/jour...
June 2, 2025 at 1:17 PM
Review of methane through space and time, coauthored with @planetdr.bsky.social, is out (OA) in @annualreviews.bsky.social.
Hopefully a helpful resource for community and courses. www.annualreviews.org/content/jour...
Hopefully a helpful resource for community and courses. www.annualreviews.org/content/jour...
Congratulations to the LVK community on this record achievement! The ton of upgrade work done since O3 really paid off 👏
🎉 We have reached 200 #GravitationalWave candidates in O4! 🎉
The fourth observing run (O4) of our detector network has had the best performance so far, with more candidates than ever before! We are currently busy analysing all these wonderful data and look forward to sharing results
#O4IsHere 🔭☄️
The fourth observing run (O4) of our detector network has had the best performance so far, with more candidates than ever before! We are currently busy analysing all these wonderful data and look forward to sharing results
#O4IsHere 🔭☄️
March 19, 2025 at 2:34 PM
Congratulations to the LVK community on this record achievement! The ton of upgrade work done since O3 really paid off 👏
Overall I think o3-mini will be a very useful model for the academic community.
Learn more here: openai.com/index/openai...
Learn more here: openai.com/index/openai...
OpenAI o3-mini
Pushing the frontier of cost-effective reasoning.
openai.com
January 31, 2025 at 7:21 PM
Overall I think o3-mini will be a very useful model for the academic community.
Learn more here: openai.com/index/openai...
Learn more here: openai.com/index/openai...
Apple: "LLMs don't really think."
AI community: "Here we go again 🙄"
But wait—they tested reasoning models on 4 puzzles. Above certain complexity? Everyone fails. Even o3.
The paper → ml-site.cdn-apple.com/papers/the-i...
AI community: "Here we go again 🙄"
But wait—they tested reasoning models on 4 puzzles. Above certain complexity? Everyone fails. Even o3.
The paper → ml-site.cdn-apple.com/papers/the-i...
ml-site.cdn-apple.com
June 8, 2025 at 7:45 PM
Apple: "LLMs don't really think."
AI community: "Here we go again 🙄"
But wait—they tested reasoning models on 4 puzzles. Above certain complexity? Everyone fails. Even o3.
The paper → ml-site.cdn-apple.com/papers/the-i...
AI community: "Here we go again 🙄"
But wait—they tested reasoning models on 4 puzzles. Above certain complexity? Everyone fails. Even o3.
The paper → ml-site.cdn-apple.com/papers/the-i...
Announcing ARBOR, an open research community for collectively understanding how reasoning models like openai-o3 and deepseek-r1 work. We invite all researchers and enthusiasts to this initiative by @wattenberg.bsky.social's and @davidbau.bsky.social's lab.
arborproject.github.io
arborproject.github.io
February 20, 2025 at 7:55 PM
Announcing ARBOR, an open research community for collectively understanding how reasoning models like openai-o3 and deepseek-r1 work. We invite all researchers and enthusiasts to this initiative by @wattenberg.bsky.social's and @davidbau.bsky.social's lab.
arborproject.github.io
arborproject.github.io
This makes me so happy to live in the San Francisco Bay Area.
www.reddit.com/r/AskSF/s/O3...
www.reddit.com/r/AskSF/s/O3...
From the AskSF community on Reddit
Explore this post and more from the AskSF community
www.reddit.com
August 28, 2025 at 2:02 AM
This makes me so happy to live in the San Francisco Bay Area.
www.reddit.com/r/AskSF/s/O3...
www.reddit.com/r/AskSF/s/O3...
Then you don’t know anything about the district, which contains a significant Jewish population descended from Holocaust survivors.
March 16, 2026 at 9:08 PM
Then you don’t know anything about the district, which contains a significant Jewish population descended from Holocaust survivors.
happy 25th birthday transgender flag 🏳️⚧️ 🏳️⚧️ 🏳️⚧️ :O3
twitter.com/25YearsAgoLi...
twitter.com/25YearsAgoLi...
March 7, 2024 at 12:55 AM
happy 25th birthday transgender flag 🏳️⚧️ 🏳️⚧️ 🏳️⚧️ :O3
twitter.com/25YearsAgoLi...
twitter.com/25YearsAgoLi...
Ich war schon sehr früh bei FB, also zu Zeiten, als alles nur Englisch war und sich nur Studenten und Uni-Mitarbeiter anmelden durften. Ich hab mich sehr gewundert, was der damals beginnende Hype soll, denn uboot.com, sms.at und selbst die ORF-/Ö3-Community konnten viel mehr.
January 20, 2025 at 9:34 PM
Bigger fish to fry in the world right now, but I continue to ask: what is wrong with the AI benchmark community that those running the benchmarks don't even have all the relevant details about their participants?
techcrunch.com/2025/04/02/o...
techcrunch.com/2025/04/02/o...
OpenAI's o3 model might be costlier to run than originally estimated | TechCrunch
OpenAI's o3 model might be costlier to run than originally estimated, according to a third-party benchmarking org.
techcrunch.com
April 3, 2025 at 2:09 PM
Bigger fish to fry in the world right now, but I continue to ask: what is wrong with the AI benchmark community that those running the benchmarks don't even have all the relevant details about their participants?
techcrunch.com/2025/04/02/o...
techcrunch.com/2025/04/02/o...
update the zine put together by @yiffmagazine.bsky.social was such a beautiful read. I feel so lucky to be a small part of a community like this. :O3
received many incredible things today
November 19, 2024 at 12:35 AM
update the zine put together by @yiffmagazine.bsky.social was such a beautiful read. I feel so lucky to be a small part of a community like this. :O3
Yep, Max Schwarzer, VP of Research / post-training at OpenAI, just left OpenAI for Anthropic.
www.reddit.com/r/ClaudeAI/s...
www.reddit.com/r/ClaudeAI/s...
From the ClaudeAI community on Reddit: OpenAI's post-training lead leaves and joins Anthropic: he helped ship GPT-5, 5.1, 5.2, 5.3-Codex, o3 and o1 and will return to hands-on RL research at Anthropic
Explore this post and more from the ClaudeAI community
www.reddit.com
March 4, 2026 at 11:32 PM
Yep, Max Schwarzer, VP of Research / post-training at OpenAI, just left OpenAI for Anthropic.
www.reddit.com/r/ClaudeAI/s...
www.reddit.com/r/ClaudeAI/s...
1/2 "They did reveal that o3 was fine-tuned on 300 out of 400 of the ARC training tasks"
This seems like a bit of a cheat to me. I'd rather it see novel problems. Why not let the community create thousands of challenges? As someone that plays with AI image/video generators, it's interesting to see..
This seems like a bit of a cheat to me. I'd rather it see novel problems. Why not let the community create thousands of challenges? As someone that plays with AI image/video generators, it's interesting to see..
December 23, 2024 at 3:21 PM
1/2 "They did reveal that o3 was fine-tuned on 300 out of 400 of the ARC training tasks"
This seems like a bit of a cheat to me. I'd rather it see novel problems. Why not let the community create thousands of challenges? As someone that plays with AI image/video generators, it's interesting to see..
This seems like a bit of a cheat to me. I'd rather it see novel problems. Why not let the community create thousands of challenges? As someone that plays with AI image/video generators, it's interesting to see..
Perplexity 無料だから試してるんだけど、こんだけ引用しまくるようなシステムでも普通にハルシネーションすることがある... R1 だからなのかなぁ。 o3-mini にしたら減るんやろか。
www.reddit.com/r/perplexity...
www.reddit.com/r/perplexity...
From the perplexity_ai community on Reddit
Explore this post and more from the perplexity_ai community
www.reddit.com
February 12, 2025 at 7:54 AM
Perplexity 無料だから試してるんだけど、こんだけ引用しまくるようなシステムでも普通にハルシネーションすることがある... R1 だからなのかなぁ。 o3-mini にしたら減るんやろか。
www.reddit.com/r/perplexity...
www.reddit.com/r/perplexity...
The development of o3 has sparked excitement and debate within the AI community.
Some see it as a clear indication of the rapid progress being made towards AGI, while others remain skeptical, emphasizing the need for further advancements and cost reductions.
Some see it as a clear indication of the rapid progress being made towards AGI, while others remain skeptical, emphasizing the need for further advancements and cost reductions.
December 21, 2024 at 2:32 AM
The development of o3 has sparked excitement and debate within the AI community.
Some see it as a clear indication of the rapid progress being made towards AGI, while others remain skeptical, emphasizing the need for further advancements and cost reductions.
Some see it as a clear indication of the rapid progress being made towards AGI, while others remain skeptical, emphasizing the need for further advancements and cost reductions.
LLMs that failed miserably in 2024!
analyticsindiamag.com/ai-trends/ll...
#XAI #LLM #LLMs #GenAI #GenerativeAI #OpenAI #ChatGPT #GPT #tech #AI #technology #artificalintelligence #Chatbot #MetaAI #llama31 #Llama3 #GROK2 #GROK2AI #GROK #chatgpt4 #Gemini #aisafety #AIethics #ethicalai #llama #AI
analyticsindiamag.com/ai-trends/ll...
#XAI #LLM #LLMs #GenAI #GenerativeAI #OpenAI #ChatGPT #GPT #tech #AI #technology #artificalintelligence #Chatbot #MetaAI #llama31 #Llama3 #GROK2 #GROK2AI #GROK #chatgpt4 #Gemini #aisafety #AIethics #ethicalai #llama #AI
LLMs that Failed Miserably in 2024
In 2024, the AI community witnessed the launch of several new large language models (LLMs), such as OpenAI’s o3 and Google Gemini 2, which promised to push the boundaries of what’s possible with AI.
analyticsindiamag.com
January 4, 2025 at 3:52 AM
Deep Researchが書いた文書を専門の人が読んだところ、mediocre（平凡、2流）だそうだ
www.reddit.com/r/OpenAI/s/V...
www.reddit.com/r/OpenAI/s/V...
From the OpenAI community on Reddit: How good is this paper by o3 Deep Research? Full PDF and source in comments.
Explore this post and more from the OpenAI community
www.reddit.com
February 3, 2025 at 7:17 AM
Deep Researchが書いた文書を専門の人が読んだところ、mediocre（平凡、2流）だそうだ
www.reddit.com/r/OpenAI/s/V...
www.reddit.com/r/OpenAI/s/V...