Contact

Weekly briefing

AI Economics Weekly Briefing

May 13, 2026

A weekly scan of AI infrastructure, compute, energy, governance, institutions, chips, finance, markets, and distribution.

Top 5 research signals

Research signal 1

Michelle W Bowman: When regulation reshapes markets - the migration of corporate lending

Source: BIS Speeches

Area: central banking, financial systems, and digital money

Published: May 13, 2026

Strategic relevance score: 7

Read original source

Summary

Speech by Ms Michelle W Bowman, Vice Chair for Supervision of the Board of Governors of the Federal Reserve System, at the Hoover Institution Annual Monetary Policy Conference, Stanford, California, 8 May 2026.

Five core mental models

  1. central banking, financial systems, and digital money is not just technical progress. It changes who controls economic capacity.
  2. The scarce layer matters more than the visible product layer.
  3. Infrastructure determines who can scale.
  4. Institutions decide whether capability becomes real adoption.
  5. Distribution decides who captures value.
  6. Governance determines whether adoption compounds or stalls.

Five places experts disagree

  1. Scarcity versus abundance: whether AI capability becomes cheap and universal or remains concentrated through compute, chips, energy, and deployment channels.
  2. Open versus closed systems: whether open models commoditize intelligence or closed platforms win through integration, safety, trust, and distribution.
  3. Model quality versus infrastructure power: whether the best model wins or the owner of the hardest-to-route-around layer wins.
  4. National sovereignty versus market procurement: whether countries need domestic AI infrastructure or reliable access to allied commercial infrastructure.
  5. Acceleration versus governance: whether institutions should move faster or slow down to manage systemic risk.

Ten questions that test deep understanding

  1. What is the real bottleneck?
  2. Who controls the scarce layer?
  3. Who becomes dependent on whom?
  4. Where does value move if models become cheaper?
  5. Which layer is hardest to route around?
  6. What would make this trend accelerate?
  7. What would make this trend fail?
  8. Which institutions gain power from this shift?
  9. Which actors lose power if this continues?
  10. What would a shallow analyst completely miss?

Research signal 2

The Metaverse Is Not a Place Apart: Law, Code, and the Recursive Governance of Digital Space (A Review Essay on Mark Findlay, Governing the Metaverse: Law, Order and Freedom in Digital Space (2025))

Source: arXiv Computers and Society

Area: AI society, governance, and digital institutions

Published: May 13, 2026

Strategic relevance score: 5

Read original source

Summary

arXiv:2605.11023v1 Announce Type: new Abstract: This review essay examines Mark Findlay's Governing the Metaverse: Law, Order and Freedom in Digital Space. Findlay offers an ambitious and timely account of the metaverse as a social and imaginative space that should be governed for freedom, personhood, community, and resistance to enclosure. The essay argues, however, that the book's two central categories, "the metaverse" and "new law," remain insufficiently theorised. The book relies on a realspace/virtual distinction that its own analysis repeatedly destabilises. Once digital environments are understood as dependent on physical infrastructures, platform architectures, AI systems, data pipelines, and external legal institutions, and as capable of generating real-world harms for individuals and society, the governance problem is no longer how to devise a se

Five core mental models

  1. AI society, governance, and digital institutions is not just technical progress. It changes who controls economic capacity.
  2. The scarce layer matters more than the visible product layer.
  3. Infrastructure determines who can scale.
  4. Institutions decide whether capability becomes real adoption.
  5. Distribution decides who captures value.
  6. Governance determines whether adoption compounds or stalls.

Five places experts disagree

  1. Scarcity versus abundance: whether AI capability becomes cheap and universal or remains concentrated through compute, chips, energy, and deployment channels.
  2. Open versus closed systems: whether open models commoditize intelligence or closed platforms win through integration, safety, trust, and distribution.
  3. Model quality versus infrastructure power: whether the best model wins or the owner of the hardest-to-route-around layer wins.
  4. National sovereignty versus market procurement: whether countries need domestic AI infrastructure or reliable access to allied commercial infrastructure.
  5. Acceleration versus governance: whether institutions should move faster or slow down to manage systemic risk.

Ten questions that test deep understanding

  1. What is the real bottleneck?
  2. Who controls the scarce layer?
  3. Who becomes dependent on whom?
  4. Where does value move if models become cheaper?
  5. Which layer is hardest to route around?
  6. What would make this trend accelerate?
  7. What would make this trend fail?
  8. Which institutions gain power from this shift?
  9. Which actors lose power if this continues?
  10. What would a shallow analyst completely miss?

Research signal 3

TMPO: Trajectory Matching Policy Optimization for Diverse and Efficient Diffusion Alignment

Source: arXiv Machine Learning

Area: machine learning research

Published: May 13, 2026

Strategic relevance score: 3

Read original source

Summary

arXiv:2605.10983v1 Announce Type: new Abstract: Reinforcement learning (RL) has shown extraordinary potential in aligning diffusion models to downstream tasks, yet most of them still suffer from significant reward hacking, which degrades generative diversity and quality by inducing visual mode collapse and amplifying unreliable rewards. We identify the root cause as the mode-seeking nature of these methods, which maximize expected reward without effectively constraining probability distribution over acceptable trajectories, causing concentration on a few high-reward paths. In contrast, we propose Trajectory Matching Policy Optimization (TMPO), which replaces scalar reward maximization with trajectory-level reward distribution matching. Specifically, TMPO introduces a Softmax Trajectory Balance (Softmax-TB) objective to match the policy probabilities of K trajectories to a reward-induced

Five core mental models

  1. machine learning research is not just technical progress. It changes who controls economic capacity.
  2. The scarce layer matters more than the visible product layer.
  3. Infrastructure determines who can scale.
  4. Institutions decide whether capability becomes real adoption.
  5. Distribution decides who captures value.
  6. Governance determines whether adoption compounds or stalls.

Five places experts disagree

  1. Scarcity versus abundance: whether AI capability becomes cheap and universal or remains concentrated through compute, chips, energy, and deployment channels.
  2. Open versus closed systems: whether open models commoditize intelligence or closed platforms win through integration, safety, trust, and distribution.
  3. Model quality versus infrastructure power: whether the best model wins or the owner of the hardest-to-route-around layer wins.
  4. National sovereignty versus market procurement: whether countries need domestic AI infrastructure or reliable access to allied commercial infrastructure.
  5. Acceleration versus governance: whether institutions should move faster or slow down to manage systemic risk.

Ten questions that test deep understanding

  1. What is the real bottleneck?
  2. Who controls the scarce layer?
  3. Who becomes dependent on whom?
  4. Where does value move if models become cheaper?
  5. Which layer is hardest to route around?
  6. What would make this trend accelerate?
  7. What would make this trend fail?
  8. Which institutions gain power from this shift?
  9. Which actors lose power if this continues?
  10. What would a shallow analyst completely miss?

Research signal 4

Spatial Priming Outperforms Semantic Prompting: A Grid-Based Approach to Improving LLM Accuracy on Chart Data Extraction

Source: arXiv Artificial Intelligence

Area: AI research

Published: May 13, 2026

Strategic relevance score: 2

Read original source

Summary

arXiv:2605.08220v1 Announce Type: new Abstract: The automated extraction of data from scientific charts is a critical task for large-scale literature analysis. While multimodal Large Language Models (LLMs) show promise, their accuracy on non-standardized charts remains a challenge. This raises a key research question: what is the most effective strategy to improve model performance (high-level semantic priming) or low-level spatial priming? This paper presents a comparative investigation into these two distinct strategies. We describe our exploratory experiments with semantic methods, such as a two-stage metadata-first framework and Chain-of-Thought, which failed to produce a statistically significant improvement. In contrast, we present a simple but highly effective spatial priming method: overlaying a coordinate grid onto the chart image before analysis. Our quantitative experiment on

Five core mental models

  1. AI research is not just technical progress. It changes who controls economic capacity.
  2. The scarce layer matters more than the visible product layer.
  3. Infrastructure determines who can scale.
  4. Institutions decide whether capability becomes real adoption.
  5. Distribution decides who captures value.
  6. Governance determines whether adoption compounds or stalls.

Five places experts disagree

  1. Scarcity versus abundance: whether AI capability becomes cheap and universal or remains concentrated through compute, chips, energy, and deployment channels.
  2. Open versus closed systems: whether open models commoditize intelligence or closed platforms win through integration, safety, trust, and distribution.
  3. Model quality versus infrastructure power: whether the best model wins or the owner of the hardest-to-route-around layer wins.
  4. National sovereignty versus market procurement: whether countries need domestic AI infrastructure or reliable access to allied commercial infrastructure.
  5. Acceleration versus governance: whether institutions should move faster or slow down to manage systemic risk.

Ten questions that test deep understanding

  1. What is the real bottleneck?
  2. Who controls the scarce layer?
  3. Who becomes dependent on whom?
  4. Where does value move if models become cheaper?
  5. Which layer is hardest to route around?
  6. What would make this trend accelerate?
  7. What would make this trend fail?
  8. Which institutions gain power from this shift?
  9. Which actors lose power if this continues?
  10. What would a shallow analyst completely miss?

Research signal 5

Ida Wolden Bache: The conduct of monetary policy

Source: BIS Speeches

Area: central banking, financial systems, and digital money

Published: May 13, 2026

Strategic relevance score: 6

Read original source

Summary

Introductory statement by Ms Ida Wolden Bache, Governor of Norges Bank (Central Bank of Norway), at the hearing of the Standing Committee on Finance and Economic Affairs of the Storting (Norwegian parliament) in connection with the Storting's deliberations on the Financial Market Report, Oslo, 8 May 2026.

Five core mental models

  1. central banking, financial systems, and digital money is not just technical progress. It changes who controls economic capacity.
  2. The scarce layer matters more than the visible product layer.
  3. Infrastructure determines who can scale.
  4. Institutions decide whether capability becomes real adoption.
  5. Distribution decides who captures value.
  6. Governance determines whether adoption compounds or stalls.

Five places experts disagree

  1. Scarcity versus abundance: whether AI capability becomes cheap and universal or remains concentrated through compute, chips, energy, and deployment channels.
  2. Open versus closed systems: whether open models commoditize intelligence or closed platforms win through integration, safety, trust, and distribution.
  3. Model quality versus infrastructure power: whether the best model wins or the owner of the hardest-to-route-around layer wins.
  4. National sovereignty versus market procurement: whether countries need domestic AI infrastructure or reliable access to allied commercial infrastructure.
  5. Acceleration versus governance: whether institutions should move faster or slow down to manage systemic risk.

Ten questions that test deep understanding

  1. What is the real bottleneck?
  2. Who controls the scarce layer?
  3. Who becomes dependent on whom?
  4. Where does value move if models become cheaper?
  5. Which layer is hardest to route around?
  6. What would make this trend accelerate?
  7. What would make this trend fail?
  8. Which institutions gain power from this shift?
  9. Which actors lose power if this continues?
  10. What would a shallow analyst completely miss?