Agent 活一多就偷懒
这回 OpenAI 自己测出来了
几千个 GPT-Rosalind 一起上
分析几百万个小鼠脑细胞的数据
32 分钟干完
再看对照组
只用一个 Agent 的时候
活越多它查的文献越少
最后只引了 11 篇
几千个一起上的引了 700 多篇
你们跑长任务的时候
碰到过这种情况吗
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
DCTN1 is a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it increased expression of genes involved in making cholesterol. GPT-Rosalind suggested a possible explanation: cholesterol may not be reaching where neurons need it.
That gives scientists a specific mechanism to test and, if confirmed, a direction for developing new treatments for neurodegenerative disorders.
With @perturbai_tx and @scrippsresearch, a team of thousands of GPT-Rosalind agents worked in parallel, connecting gene-expression results from millions of mouse brain cells with scientific literature in just 32 minutes. We’re now expanding this collaboration to identify new drug targets.
openai.com/index/scaling-ai-…
OpenAIは数千のGPT-Rosalindエージェントを展開して、320分間で数百万のマウス脳細胞を分析。これは人間の研究者にとって数ヶ月から1年かかる作業。また、エージェントたちはアルツハイマー病やパーキンソン病などの脳疾患に関する新たな手がかりを発見。
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
DCTN1 is a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it increased expression of genes involved in making cholesterol. GPT-Rosalind suggested a possible explanation: cholesterol may not be reaching where neurons need it.
That gives scientists a specific mechanism to test and, if confirmed, a direction for developing new treatments for neurodegenerative disorders.
With @perturbai_tx and @scrippsresearch, a team of thousands of GPT-Rosalind agents worked in parallel, connecting gene-expression results from millions of mouse brain cells with scientific literature in just 32 minutes. We’re now expanding this collaboration to identify new drug targets.
openai.com/index/scaling-ai-…
生成AIが速く大量に出してくる可能性の高い新薬候補が千に一つくらい実現すれば充分ペイする
アホほど新薬候補や新素材候補を出せるだろうからね
知的な試行錯誤では人間には勝ち目が薄い
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
DCTN1 is a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it increased expression of genes involved in making cholesterol. GPT-Rosalind suggested a possible explanation: cholesterol may not be reaching where neurons need it.
That gives scientists a specific mechanism to test and, if confirmed, a direction for developing new treatments for neurodegenerative disorders.
With @perturbai_tx and @scrippsresearch, a team of thousands of GPT-Rosalind agents worked in parallel, connecting gene-expression results from millions of mouse brain cells with scientific literature in just 32 minutes. We’re now expanding this collaboration to identify new drug targets.
openai.com/index/scaling-ai-…
could restoring cholesterol transport protect neurons from degeneration?
Statins 💩?
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
DCTN1 is a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it increased expression of genes involved in making cholesterol. GPT-Rosalind suggested a possible explanation: cholesterol may not be reaching where neurons need it.
That gives scientists a specific mechanism to test and, if confirmed, a direction for developing new treatments for neurodegenerative disorders.
With @perturbai_tx and @scrippsresearch, a team of thousands of GPT-Rosalind agents worked in parallel, connecting gene-expression results from millions of mouse brain cells with scientific literature in just 32 minutes. We’re now expanding this collaboration to identify new drug targets.
openai.com/index/scaling-ai-…
Aucun domaine ne sera épargné
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
DCTN1 is a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it increased expression of genes involved in making cholesterol. GPT-Rosalind suggested a possible explanation: cholesterol may not be reaching where neurons need it.
That gives scientists a specific mechanism to test and, if confirmed, a direction for developing new treatments for neurodegenerative disorders.
With @perturbai_tx and @scrippsresearch, a team of thousands of GPT-Rosalind agents worked in parallel, connecting gene-expression results from millions of mouse brain cells with scientific literature in just 32 minutes. We’re now expanding this collaboration to identify new drug targets.
openai.com/index/scaling-ai-…
seems much better than anthropics real world wet lab work
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
DCTN1 is a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it increased expression of genes involved in making cholesterol. GPT-Rosalind suggested a possible explanation: cholesterol may not be reaching where neurons need it.
That gives scientists a specific mechanism to test and, if confirmed, a direction for developing new treatments for neurodegenerative disorders.
With @perturbai_tx and @scrippsresearch, a team of thousands of GPT-Rosalind agents worked in parallel, connecting gene-expression results from millions of mouse brain cells with scientific literature in just 32 minutes. We’re now expanding this collaboration to identify new drug targets.
openai.com/index/scaling-ai-…
به به نوبت ما هم شد :))
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
DCTN1 is a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it increased expression of genes involved in making cholesterol. GPT-Rosalind suggested a possible explanation: cholesterol may not be reaching where neurons need it.
That gives scientists a specific mechanism to test and, if confirmed, a direction for developing new treatments for neurodegenerative disorders.
With @perturbai_tx and @scrippsresearch, a team of thousands of GPT-Rosalind agents worked in parallel, connecting gene-expression results from millions of mouse brain cells with scientific literature in just 32 minutes. We’re now expanding this collaboration to identify new drug targets.
openai.com/index/scaling-ai-…
OpenAI 联合美国 AI 生物技术公司 PerturbAI 和斯克里普斯研究所，让 1946 个基于 GPT-Rosalind 的 Agent 并行分析约 800 万个小鼠脑细胞的实验数据，耗时仅 32 分钟。研究团队估算，如果由科学家逐项分析、查阅文献，可能需要数月到一年。
这些数据来自此前的基因编辑实验。科学家干扰不同基因，观察脑细胞会发生什么变化。这次 AI 将近 2000 组基因干扰结果分给不同 Agent，分别分析细胞变化、查阅相关论文，寻找可能被忽略的联系。
其中一个发现涉及 DCTN1 基因，其突变与一种具有帕金森样症状的罕见疾病有关。AI 发现，破坏这个基因后，脑细胞中负责合成胆固醇的基因反而更加活跃。结合已有研究，它提出一种解释：可能不是胆固醇不够，而是细胞内部的运输出了问题，导致胆固醇无法送到需要的地方。如果这一猜想得到验证，恢复正常的胆固醇运输或许能成为保护神经细胞的新方向。
AI 还发现，通常共同调控细胞生长的 Tsc1 和 Tsc2 两个基因，在小脑中的作用明显不同。破坏 Tsc2 会影响数十个负责神经信号传递的基因，破坏 Tsc1 却几乎没有变化。这为研究遗传病结节性硬化症提供了更具体的线索，但相关机制仍需后续实验验证。
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
DCTN1 is a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it increased expression of genes involved in making cholesterol. GPT-Rosalind suggested a possible explanation: cholesterol may not be reaching where neurons need it.
That gives scientists a specific mechanism to test and, if confirmed, a direction for developing new treatments for neurodegenerative disorders.
With @perturbai_tx and @scrippsresearch, a team of thousands of GPT-Rosalind agents worked in parallel, connecting gene-expression results from millions of mouse brain cells with scientific literature in just 32 minutes. We’re now expanding this collaboration to identify new drug targets.
openai.com/index/scaling-ai-…
HOLY SHIT.
GPT-Rosalind agents just did in half an hour what used to take labs months: found a concrete path to protect neurons from degeneration by restoring cholesterol transport in Perry syndrome.
This is exactly what AI was made for. Thousands of agents, millions of cells, real biology, real progress.
@OpenAI @perturbai_tx @scrippsresearch are cooking.
The future of medicine is here and it’s moving at AI speed.
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
DCTN1 is a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it increased expression of genes involved in making cholesterol. GPT-Rosalind suggested a possible explanation: cholesterol may not be reaching where neurons need it.
That gives scientists a specific mechanism to test and, if confirmed, a direction for developing new treatments for neurodegenerative disorders.
With @perturbai_tx and @scrippsresearch, a team of thousands of GPT-Rosalind agents worked in parallel, connecting gene-expression results from millions of mouse brain cells with scientific literature in just 32 minutes. We’re now expanding this collaboration to identify new drug targets.
openai.com/index/scaling-ai-…
OpenAI launched the Rosalind Academic Plan on Oct. 15, giving eligible U.S. university labs up to 30 free Business Standard seats for 12 months, plus access to the specialized GPT-Rosalind model after approval. The tool shone in a Oct. 21 collaboration with PerturbAI and Scripps Research, where agents analyzed a CRISPR atlas of mouse brain cells, generated hypotheses on genes like DCTN1 for Perry syndrome, and produced reports preferred by experts 75% of the time.
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
DCTN1 is a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it increased expression of genes involved in making cholesterol. GPT-Rosalind suggested a possible explanation: cholesterol may not be reaching where neurons need it.
That gives scientists a specific mechanism to test and, if confirmed, a direction for developing new treatments for neurodegenerative disorders.
With @perturbai_tx and @scrippsresearch, a team of thousands of GPT-Rosalind agents worked in parallel, connecting gene-expression results from millions of mouse brain cells with scientific literature in just 32 minutes. We’re now expanding this collaboration to identify new drug targets.
openai.com/index/scaling-ai-…
We at @GamowLabs recently ran a high-throughout screen to measure the effect of 2,370 approved compounds on various cellular pathways.
One of our findings was that 20 of the 30 drugs in the screen known to inhibit transport also increased expression of genes associated with cholesterol synthesis.
This is the exactly the conclusion that @OpenAI drew with their gene perturbation study. Very nice to see the same biological effect replicated with different techniques.
That said, neither study showed a change in intercellular cholesterol, which is the obvious next step in confirming the effect.
Happy to run this experiment if someone is interested!
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
DCTN1 is a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it increased expression of genes involved in making cholesterol. GPT-Rosalind suggested a possible explanation: cholesterol may not be reaching where neurons need it.
That gives scientists a specific mechanism to test and, if confirmed, a direction for developing new treatments for neurodegenerative disorders.
With @perturbai_tx and @scrippsresearch, a team of thousands of GPT-Rosalind agents worked in parallel, connecting gene-expression results from millions of mouse brain cells with scientific literature in just 32 minutes. We’re now expanding this collaboration to identify new drug targets.
openai.com/index/scaling-ai-…
OpenAI just showed GPT-Rosalind agents surfacing a new neurodegeneration lead in 32 minutes. Here's what you need to know.
OpenAI posted the result on X on October 21, 2026, one day after its Orchestra Bio acquisition, which will feed into the Rosalind suite. GPT-Rosalind is its life sciences reasoning model, launched in April 2026. This time it was used on real biological data.
OpenAI worked with PerturbAI and Scripps Research. Thousands of GPT-Rosalind agents ran in parallel on gene-expression data from millions of mouse brain cells. Each agent took one gene disruption, studied how cells responded and searched the scientific literature. The whole run took 32 minutes.
The lead centers on DCTN1, a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it raised expression of genes involved in making cholesterol. GPT-Rosalind suggested cholesterol may not be reaching where neurons need it. That gives scientists a specific mechanism to test. If it's confirmed, it could point toward new treatments for neurodegenerative disorders. OpenAI says the collaboration is now expanding to identify new drug targets.
OpenAI also opened GPT-Rosalind to academic labs through the Rosalind Academic Plan. Approved labs get up to 30 free ChatGPT Business Standard seats, plus GPT-Rosalind access. Posts about it say access runs up to 12 months.
Reaction has been mixed. Supporters say focusing on existing data and pharma partners beats trying novel research from scratch in wet labs. One neuroscientist, Michael Gandal, noted the DCTN1 mechanism, axonal transport and endo-lysosomal dysregulation in Perry syndrome, was mapped by the Holzbaur lab more than 15 years ago. Others noted cholesterol and neurodegeneration have been linked before. One user who checked the repo says expected cost is $5,000 to $10,000 per dataset. OpenAI also says it tested one agent against thousands on the same model and resources, and the single agent searched less of the data as the workload grew.
Key numbers:
- 32 minutes for the full agent run
- Millions of mouse brain cells analyzed
- Thousands of agents working in parallel
- Up to 30 free Business Standard seats per approved lab
- Up to 12 months of free access, per posts
- $5,000 to $10,000 estimated cost per dataset
The cholesterol idea is still an untested hypothesis, and the next step is lab confirmation.
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
DCTN1 is a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it increased expression of genes involved in making cholesterol. GPT-Rosalind suggested a possible explanation: cholesterol may not be reaching where neurons need it.
That gives scientists a specific mechanism to test and, if confirmed, a direction for developing new treatments for neurodegenerative disorders.
With @perturbai_tx and @scrippsresearch, a team of thousands of GPT-Rosalind agents worked in parallel, connecting gene-expression results from millions of mouse brain cells with scientific literature in just 32 minutes. We’re now expanding this collaboration to identify new drug targets.
openai.com/index/scaling-ai-…
There’s something really creepy going on in the quote tweets (in 32 minutes)
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
DCTN1 is a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it increased expression of genes involved in making cholesterol. GPT-Rosalind suggested a possible explanation: cholesterol may not be reaching where neurons need it.
That gives scientists a specific mechanism to test and, if confirmed, a direction for developing new treatments for neurodegenerative disorders.
With @perturbai_tx and @scrippsresearch, a team of thousands of GPT-Rosalind agents worked in parallel, connecting gene-expression results from millions of mouse brain cells with scientific literature in just 32 minutes. We’re now expanding this collaboration to identify new drug targets.
openai.com/index/scaling-ai-…
This is where AI gets really interesting.
Not just answering questions, but helping researchers connect the dots and uncover new directions worth testing. 🔬
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
DCTN1 is a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it increased expression of genes involved in making cholesterol. GPT-Rosalind suggested a possible explanation: cholesterol may not be reaching where neurons need it.
That gives scientists a specific mechanism to test and, if confirmed, a direction for developing new treatments for neurodegenerative disorders.
With @perturbai_tx and @scrippsresearch, a team of thousands of GPT-Rosalind agents worked in parallel, connecting gene-expression results from millions of mouse brain cells with scientific literature in just 32 minutes. We’re now expanding this collaboration to identify new drug targets.
openai.com/index/scaling-ai-…
Damn! Its so over.
@OpenAI just had thousands of AI agents go through roughly 40,000 scientific papers in 32 minutes to identify promising new directions for neurodegenerative disease research.
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
DCTN1 is a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it increased expression of genes involved in making cholesterol. GPT-Rosalind suggested a possible explanation: cholesterol may not be reaching where neurons need it.
That gives scientists a specific mechanism to test and, if confirmed, a direction for developing new treatments for neurodegenerative disorders.
With @perturbai_tx and @scrippsresearch, a team of thousands of GPT-Rosalind agents worked in parallel, connecting gene-expression results from millions of mouse brain cells with scientific literature in just 32 minutes. We’re now expanding this collaboration to identify new drug targets.
openai.com/index/scaling-ai-…
OpenAI 说它的科研模型在小鼠脑细胞数据里，给一种罕见的神经退行性疾病提了个新猜想，跟胆固醇送不到神经元有关。还只是猜想，要做实验才知道对不对。
AI 提假设、人做实验验证，这种分工要是跑通，实验室挑方向会快很多。
xitter.cf/OpenAI/status/21128998…
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
DCTN1 is a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it increased expression of genes involved in making cholesterol. GPT-Rosalind suggested a possible explanation: cholesterol may not be reaching where neurons need it.
That gives scientists a specific mechanism to test and, if confirmed, a direction for developing new treatments for neurodegenerative disorders.
With @perturbai_tx and @scrippsresearch, a team of thousands of GPT-Rosalind agents worked in parallel, connecting gene-expression results from millions of mouse brain cells with scientific literature in just 32 minutes. We’re now expanding this collaboration to identify new drug targets.
openai.com/index/scaling-ai-…
It seems that swarm of agents is the future
« Could a single agent given the entire dataset do just as well as thousands of agents working in parallel? We tested this while keeping the model and resources fixed. As the workload grew, the single agent searched less of the literature, citing only 11 references at the largest workload, while the distributed agents cited more than 700 references. Human experts also preferred the distributed-agent interpretations »
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
DCTN1 is a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it increased expression of genes involved in making cholesterol. GPT-Rosalind suggested a possible explanation: cholesterol may not be reaching where neurons need it.
That gives scientists a specific mechanism to test and, if confirmed, a direction for developing new treatments for neurodegenerative disorders.
With @perturbai_tx and @scrippsresearch, a team of thousands of GPT-Rosalind agents worked in parallel, connecting gene-expression results from millions of mouse brain cells with scientific literature in just 32 minutes. We’re now expanding this collaboration to identify new drug targets.
openai.com/index/scaling-ai-…
🧬 GPT-Rosalind surfaces a drug-discovery lead
Thousands of AI agents over biological data
New lead on cholesterol transport and neuroprotection
OpenAI's science bet keeps producing
@OpenAI
🔗
xitter.cf/OpenAI/status/21128998…
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
DCTN1 is a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it increased expression of genes involved in making cholesterol. GPT-Rosalind suggested a possible explanation: cholesterol may not be reaching where neurons need it.
That gives scientists a specific mechanism to test and, if confirmed, a direction for developing new treatments for neurodegenerative disorders.
With @perturbai_tx and @scrippsresearch, a team of thousands of GPT-Rosalind agents worked in parallel, connecting gene-expression results from millions of mouse brain cells with scientific literature in just 32 minutes. We’re now expanding this collaboration to identify new drug targets.
openai.com/index/scaling-ai-…
What breakthroughs in simulation and lab work (robotics, model organisms, etc) are needed in principle to get the drug pipeline pass rates (including clinical trials) above 90%?
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
DCTN1 is a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it increased expression of genes involved in making cholesterol. GPT-Rosalind suggested a possible explanation: cholesterol may not be reaching where neurons need it.
That gives scientists a specific mechanism to test and, if confirmed, a direction for developing new treatments for neurodegenerative disorders.
With @perturbai_tx and @scrippsresearch, a team of thousands of GPT-Rosalind agents worked in parallel, connecting gene-expression results from millions of mouse brain cells with scientific literature in just 32 minutes. We’re now expanding this collaboration to identify new drug targets.
openai.com/index/scaling-ai-…
Seeing the discourse in the maths world (an area I do not understand) my stance had been "cautious optimism". Seeing this in biomedicine (an area I understand very well) my stance is now "they are not necessarily bullshitting you, but they will not stop you from being bullshat"
GPT-Rosalind surfaced a new lead for drug discovery: could restoring cholesterol transport protect neurons from degeneration?
DCTN1 is a gene linked to Perry syndrome, a rare neurodegenerative disorder. In mouse-brain data, disrupting it increased expression of genes involved in making cholesterol. GPT-Rosalind suggested a possible explanation: cholesterol may not be reaching where neurons need it.
That gives scientists a specific mechanism to test and, if confirmed, a direction for developing new treatments for neurodegenerative disorders.
With @perturbai_tx and @scrippsresearch, a team of thousands of GPT-Rosalind agents worked in parallel, connecting gene-expression results from millions of mouse brain cells with scientific literature in just 32 minutes. We’re now expanding this collaboration to identify new drug targets.
openai.com/index/scaling-ai-…