PSI is in the right direction. Kotar, K., Lee, W., Venkatesh, R., Chen, H., Bear, D., Watrous, J., ... & Yamins, D. (2025). World Modeling with Probabilistic Structure Integration. arXiv preprint arXiv:2509.09737.

Anything that resembles backcasting is going to be at least good. This, however, is great. arxiv.org/abs/2509.06160

LLM's are just like us! Abdelsalam, Y., Peitek, N., Maurer, A. M., Toneva, M., & Apel, S. (2025). How do Humans and LLMs Process Confusing Code?. arXiv preprint arXiv:2508.18547.

In tomorrows newsletter we'll cover: Joseph, J., & Sengul, M. (2025). Organization design: Current insights and future research directions. Journal of Management, 51(1), 249-308.

Thoroughly enjoyed this one. What a banger!

If it's a bubble, it's a bubble. Great! Focus on value? That's boring too? Back to being excited and stoked I suppose?

Wisdom here: Frey, S., Zhong, Q., Bulat, B., Weisman, W. D., Liu, C., Fujimoto, S., ... & Schweik, C. M. (2022). Governing online goods: Maturity and formalization in Minecraft, Reddit, and World of Warcraft communities. Proceedings of the ACM on Human-Computer Interaction, 6(CSCW2), 1-23.

Our featured article: Predicting gene expression from DNA sequence using deep learning models go.nature.com/3VXGCG9

This point by @earlkmiller.bsky.social is very much obvious from a mainstream deep learning perspective. Why would neurons specialize?

Alibaba launched Tongyi DeepResearch, the open-source web agent to achieve similar performance as OpenAI's Deep Research with only 30B (Activated 3B) parameters!

So cool that they did this.

Do we blow bubbles because we're bored? #AI #GAI #AGI #YOLO

If your base model is unable to solve a task, can you simply apply reinforcement learning to fix it? The answer is no. For reinforcement learning to work effectively with language models, the base model must already assign a non-negligible probability to producing a correct solution.

KHOSLA: “.. This isn’t your usual bubble mania. Most valuations are bonkers .. “.. capital concentration is extreme. .. there will only be a few winners.” 👀 @theinformation.com www.theinformation.com/articles/ai-...

This has stuck with me. It's the incentives. It always is..... ? Kalai, A. T., Nachum, O., Vempala, S. S., & Zhang, E. (2025). Why Language Models Hallucinate.

In this week's gatodo newsletter. Every Tuesday. 10AM ET.

It's just so .... weird, right? Chen, R., Arditi, A., Sleight, H., Evans, O., & Lindsey, J. (2025). Persona Vectors: Monitoring and Controlling Character Traits in Language Models. arXiv preprint arXiv:2507.21509.

MORGAN STANLEY: “.. we expect Apple to (modestly) raise iPhone prices for the first time in 7 years.” $AAPL

CHALLENGER: “.. many of these [pharma] companies are also implementing AI, causing restructuring and leading to job loss.”

Chen, R., Arditi, A., Sleight, H., Evans, O., & Lindsey, J. (2025). Persona Vectors: Monitoring and Controlling Character Traits in Language Models. arXiv preprint arXiv:2507.21509.

The makers of AI models such as OpenAI and Anthropic have discussed deals to acquire or license data from coding startups like Cursor. The data on how engineers use coding assistants could help train large models. Learn more: https://bit.ly/47kz87j

I enjoy Levin. I enjoy divergent ideas. The idea that there's something embedded in the Wiener (you're a Wiener baby!) process is attractive. Why should anything resist human understanding?

In this week's footnote: Yeah, it's a bubble. We can see that. Isn't it great that data quality is back?

Been thinking about this a lot. Shao, Y., Zope, H., Jiang, Y., Pei, J., Nguyen, D., Brynjolfsson, E., & Yang, D. (2025). Future of Work with AI Agents: Auditing Automation and Augmentation Potential across the US Workforce. arXiv preprint arXiv:2506.06576

Still love that word: HARKing.

I like small. Good direction.

Makes sense why some would call it a cult. Or a religion. Some people literally believe they're building an AI Godhead. There are a few LARPers in there, sure. But there are some folk who are complete, total, unironic, believers.

Paper: arxiv.org/abs/2508.15706 Github: github.com/tplr-ai/Spar...

Is it really the case...just do a lot of stupid things at random until you learn better? Could it really be that simple?

IDK on the GoodReads sponsorship bias. On the one hand, some readers hire publishers as a parent brand for quality (consciously or not). The tactic of sponsored reviews is an amplification of that filtering role. I get the counter-argument. I think.

Hu, Y., Diesner, J., Underwood, T., LeBlanc, Z., Layne-Worthey, G., & Downie, J. S. (2025). Who decides what is read on Goodreads? Uncovering sponsorship and its implications for scholarly Research. Big Data & Society, 12(3).

Positron is kind of neat, should make R a bit more accessible, usable?

HARKing is an amazing gift. What an acronym. Thank you. Kazinnik, S., & Brynjolfsson, E. (2025). AI and the Fed (No. w33998). National Bureau of Economic Research.

There is so much I like about CLIO. I'm ever so curious about centaurs. I don't want to eliminate the human from decision making. I'd like to liberate them make a greater impact. Obligatory #babylon5 reference to Centauri?

AI Agenda: The Power of Defaults: Why OpenAI, Anthropic and Google Are Courting App Developers Model makers are offering discounted pricing, case studies and more if customers make them the model default in their apps. Read more from @stephpalazzolo.bsky.social 👇 https://bit.ly/4mIN9k9

Backcasting truly is a superpower.

It's interesting how they calculated augmentation potential. Kazinnik, S., & Brynjolfsson, E. (2025). AI and the Fed (No. w33998). National Bureau of Economic Research.

Good morning! This week in the newsletter we're covering: Zheng, C., Liu, S., Li, M., Chen, X. H., Yu, B., Gao, C., ... & Lin, J. (2025). Group Sequence Policy Optimization. arXiv preprint arXiv:2507.18071.

TD COWEN: “.. Recent Gallup survey data shows alcohol consumption at an all-time low .. “.. supports our view that weakening consumer demand stems from a structural change in consumer perception as opposed to just cyclical economic pressures.”

Markets are inherently redistributive, right? It's just in aggregate...since 1970.... From: Fulford, S. L., & Schiantarelli, F. (2024). Still Growing Together?: The Spatial Distribution and Industrial Composition of US County GDP Since 1870. Boston College.

This is the context that we're dumping AI.... Must it be so?

From: Fulford, S. L., & Schiantarelli, F. (2024). Still Growing Together?: The Spatial Distribution and Industrial Composition of US County GDP Since 1870. Boston College.

From: Fulford, S. L., & Schiantarelli, F. (2024). Still Growing Together?: The Spatial Distribution and Industrial Composition of US County GDP Since 1870. Boston College.

From: Fulford, S. L., & Schiantarelli, F. (2024). Still Growing Together?: The Spatial Distribution and Industrial Composition of US County GDP Since 1870. Boston College.

Andrew Ng’s piece on 🇺🇸 vs 🇨🇳 competition in AI worth reading: Full article: www.deeplearning.ai/the-batch/is...

So many gains just getting left on the table rn. This is a known known. Qiu, Y., Zhao, C., Zhang, H., Zhuo, J., Li, T., Zhang, X., ... & Yang, W. Y. (2022, October). Pre-training tasks for user intent detection and embedding retrieval in e-commerce search.

Nvidia CEO's Daughter and Son Are Company's Quiet Rising Stars Read more from Anissa Gardizy 👇 www.theinformation.com/articles/nvi...

📊 AI is transforming tech workflows—67% of orgs say so. The smartest aren’t replacing talent. They’re retraining it. 🔹 Upskilling = 38% faster than hiring 🔹 Boosts retention & innovation 📥 training.linuxfoundation.org/2025-state-o... #CTO #CIO #AIimplementation #HRStrategy #Cybersecurity #LandD

Glad that we're still talking about human intelligence.