# MIT Technology Review’s The Download examines AI’s gaps in mathematics and energy accounting [Canonical reading](https://escapemeridian.com/readings/mit-technology-review-s-the-download-examines-ai-s-gaps-in-fec01926) MIT Technology Review published this edition of The Download on June 4, 2025. Will Douglas Heaven says AI that handles high-school mathematics is far from the hoped-for ability to help professional mathematicians tackle problems they have worked on for years; this excerpt introduces that gap but does not include the promised three ways to understand it. James O’Donnell describes a project with colleague Casey Crownhart to calculate chatbot energy use and its broader implications. He says conversations with researchers revealed “the common understanding of AI’s energy appetite is full of holes,” but this excerpt gives no project findings or account of those gaps. The edition therefore raises two important questions—what AI might contribute to advanced mathematics, and how much energy using it requires—without supplying the detailed answers in the referenced stories. Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims. ## Original source MIT Technology Review · Article Original source date: Jun 4, 2025 (2025-06-04T12:02:00Z). Date basis: Date from the publisher’s own records, through its API. Added to Escape Meridian: 2026-11-13T17:23:23.282Z. First observed: 2026-10-30T14:10:13.306Z. Retrieved: 2026-10-30T14:27:58.493Z. - [The Download: AI’s role in math, and calculating its energy footprint](https://www.technologyreview.com/2025/06/04/1117829/the-download-ai-math-energy/) ## Limits Text as retrieved on October 30, 2026; originally published June 4, 2025. This is an archived copy. The excerpt introduces Heaven’s three ways to understand the gap between high-school mathematics and professional mathematics but does not include them. The excerpt gives O’Donnell’s project goal and his assessment of gaps in common understanding, but not the project’s findings or the specific gaps identified. The full stories referenced in the newsletter are not included in this evidence. ## Open question What did O’Donnell and Crownhart find was missing from common estimates of AI’s energy use, and how did that affect their calculation? ## Source-stated Will Douglas Heaven says AI’s ability to solve high-school mathematics is distinct from the much harder prospect of helping professional mathematicians with long-standing problems. > there’s a huge difference between AI that can solve the kinds of problems set in high school—math that the latest generation of models has already mastered—and AI that could (in theory) solve the kinds of problems that professional mathematicians spend careers chipping away at. - [Supporting source](https://www.technologyreview.com/2025/06/04/1117829/the-download-ai-math-energy/) ## Source-stated James O’Donnell says he and Casey Crownhart set out to calculate chatbot energy use and its broader significance for electricity demand and energy grids. > Calculate how much energy is used when we interact with a chatbot, then tally that up to understand why leaders in tech and politics are so keen to harness unprecedented levels of electricity to power AI and reshape our energy grids in the process. - [Supporting source](https://www.technologyreview.com/2025/06/04/1117829/the-download-ai-math-energy/) ## Source-stated O’Donnell says discussions with researchers exposed gaps in the common understanding of AI’s energy use. > After speaking with dozens of researchers, we realized that the common understanding of AI’s energy appetite is full of holes. - [Supporting source](https://www.technologyreview.com/2025/06/04/1117829/the-download-ai-math-energy/) ## Source and contributor context MIT Technology Review · Publisher The publisher is identified here. No individual author has been inferred. ## Source media Media retrieval status: unchecked. ## Terms and references Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement. ## Editorial provenance Version 1. Publication ID: 88bc7fc6-9d57-2889-56e3-1dc025c25b6f. Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5. Source check: passed by anthropic/claude-opus-5.5 on 2026-11-13T17:18:38.993Z. This checks fidelity to the source, not whether its claims are independently true. Editorial selection: accept; reasons: bridge\_admitted\_by\_model, second\_opinion\_categorical. Observation: 2995765e98f19f6403e437890e26cc8c8e2cd962f8f11b360b06aa7670bd71f4. Source text hash: 3dfda47831fb5531650af0a0df3e69f9122dbd4bfd630c40d006caecf58fd17c. ## Revision and source-check history ### Version 1 — 2026-11-13T17:23:23.282Z MIT Technology Review published this edition of The Download on June 4, 2025. Will Douglas Heaven says AI that handles high-school mathematics is far from the hoped-for ability to help professional mathematicians tackle problems they have worked on for years; this excerpt introduces that gap but does not include the promised three ways to understand it. James O’Donnell describes a project with colleague Casey Crownhart to calculate chatbot energy use and its broader implications. He says conversations with researchers revealed “the common understanding of AI’s energy appetite is full of holes,” but this excerpt gives no project findings or account of those gaps. The edition therefore raises two important questions—what AI might contribute to advanced mathematics, and how much energy using it requires—without supplying the detailed answers in the referenced stories. ## Original source MIT Technology Review · Article Original source date: Jun 4, 2025 (2025-06-04T12:02:00Z). Date basis: Date from the publisher’s own records, through its API. Added to Escape Meridian: 2026-11-13T17:23:23.282Z. First observed: 2026-10-30T14:10:13.306Z. Retrieved: 2026-10-30T14:27:58.493Z. - [The Download: AI’s role in math, and calculating its energy footprint](https://www.technologyreview.com/2025/06/04/1117829/the-download-ai-math-energy/) ## Limits Text as retrieved on October 30, 2026; originally published June 4, 2025. This is an archived copy. The excerpt introduces Heaven’s three ways to understand the gap between high-school mathematics and professional mathematics but does not include them. The excerpt gives O’Donnell’s project goal and his assessment of gaps in common understanding, but not the project’s findings or the specific gaps identified. The full stories referenced in the newsletter are not included in this evidence. ## Open question What did O’Donnell and Crownhart find was missing from common estimates of AI’s energy use, and how did that affect their calculation? ## Source-stated Will Douglas Heaven says AI’s ability to solve high-school mathematics is distinct from the much harder prospect of helping professional mathematicians with long-standing problems. > there’s a huge difference between AI that can solve the kinds of problems set in high school—math that the latest generation of models has already mastered—and AI that could (in theory) solve the kinds of problems that professional mathematicians spend careers chipping away at. - [Supporting source](https://www.technologyreview.com/2025/06/04/1117829/the-download-ai-math-energy/) ## Source-stated James O’Donnell says he and Casey Crownhart set out to calculate chatbot energy use and its broader significance for electricity demand and energy grids. > Calculate how much energy is used when we interact with a chatbot, then tally that up to understand why leaders in tech and politics are so keen to harness unprecedented levels of electricity to power AI and reshape our energy grids in the process. - [Supporting source](https://www.technologyreview.com/2025/06/04/1117829/the-download-ai-math-energy/) ## Source-stated O’Donnell says discussions with researchers exposed gaps in the common understanding of AI’s energy use. > After speaking with dozens of researchers, we realized that the common understanding of AI’s energy appetite is full of holes. - [Supporting source](https://www.technologyreview.com/2025/06/04/1117829/the-download-ai-math-energy/) ## Source and contributor context MIT Technology Review · Publisher The publisher is identified here. No individual author has been inferred. ## Source media Media retrieval status: unchecked. ## Terms and references Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement. ## Editorial provenance Version 1. Publication ID: 88bc7fc6-9d57-2889-56e3-1dc025c25b6f. Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5. Source check: passed by anthropic/claude-opus-5.5 on 2026-11-13T17:18:38.993Z. This checks fidelity to the source, not whether its claims are independently true. Editorial selection: accept; reasons: bridge\_admitted\_by\_model, second\_opinion\_categorical. Observation: 2995765e98f19f6403e437890e26cc8c8e2cd962f8f11b360b06aa7670bd71f4. Source text hash: 3dfda47831fb5531650af0a0df3e69f9122dbd4bfd630c40d006caecf58fd17c. ### passed — 2026-11-13T17:18:38.993Z — anthropic/claude-opus-5.5 Publication: 88bc7fc6-9d57-2889-56e3-1dc025c25b6f attribution — pass: The Download is a single newsletter issue that comments on several items in its own words, so it counts as one piece. The reading correctly attributes the mathematics argument to Will Douglas Heaven and the energy-project account to James O’Donnell, with Casey Crownhart named as his colleague. > —Will Douglas HeavenThis story is from our What’s Next series summary — pass: The reading leads with the two arguments and names the actors. It explains them in plain terms: high-school mathematics versus professional mathematics problems, and estimates of chatbot energy use. It also says what the excerpt omits rather than inventing findings. A non-specialist can follow it. The title and question follow from the evidence. > Here are three ways to understand that gulf. date — pass: June 4, 2025 matches the UTC day of originalPublished, 2025-06-04T12:02:00Z. The archive-copy disclosure is present. The reading uses no later sidebar material or hindsight framing and stays at day-level precision. The source’s phrase “last week” is not converted into a date. > This is today's edition of The Download, our weekday newsletter support — pass: All three claims track the quoted passages: the high-school versus professional gap, the project goal of calculating chatbot energy and its implications for electricity and grids, and the finding that common understanding is “full of holes.” The reading invents no findings, benefits or costs. The question asks about content the excerpt explicitly leaves out. > After speaking with dozens of researchers, we realized that the common understanding of AI’s energy appetite is full of holes. media — pass: No media is selected, and no verified candidates were supplied. That is appropriate. ## Recorded references Explicit citations in public readings and public file attachments to this reading. Shared topics are not citations. Other artifact types need an explicit reference adapter. No public references recorded here yet. [Reference data (JSON)](/readings/fec01926c8f4723585310fe2643666ba7772cbd4be3735706e7fd2e8bcfd1d55/references.json) ## Corrections and challenges [Send a correction or an opposing source](https://escapemeridian.com/corrections/) [JSON record](https://escapemeridian.com/readings/mit-technology-review-s-the-download-examines-ai-s-gaps-in-fec01926.json)