MIT Technology Review · Jul 3, 2025
MIT Technology Review’s The Download warns against AI-agent hype and examines Google’s power demand
Markdown · JSON · Original source
Dated July third, twenty twenty-five, MIT Technology Review’s The Download presents two technology concerns: expectations for and the electricity used by data centers. Yoav Shoham argues that software agents could handle tasks such as booking flights and filing expenses, but warns that hype racing ahead of reality could provoke a backlash before the technology delivers real benefits. The newsletter also reports that Google signed an agreement to buy electricity from a forthcoming fusion power plant and that its data-center energy use has doubled since 2020. Casey Crownhart’s climate-newsletter item says the announcements show big tech companies hunting for clean electricity as demand and emissions grow. It also points to a key uncertainty: Google does not break out how much of its pollution is attributable to AI. The issue links enthusiasm about what AI agents might do with the practical question of how their infrastructure will be powered; it does not establish how much AI contributes to Google’s emissions.
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
Source details
Original source date: Jul 3, 2025 (2025-07-03T12:10:00Z).
Date basis: Date from the publisher’s own records, through its API.
Added to Escape Meridian: 2026-11-13T06:03:24.987Z. First observed: 2026-11-01T03:47:52.811Z. Retrieved: 2026-11-01T03:49:26.809Z.
Referenced by
Recorded citations and public file attachments. A reference alone does not establish agreement or influence.
No public references recorded here yet.
Limits
Text as retrieved on November 1, 2026; originally published July 3, 2025. The retrieved page combines newsletter items and links to fuller stories; it does not provide the full context of Shoham’s argument or further detail on Google’s electricity agreement.
Open question
How much of Google’s pollution is attributable to AI, given that the company does not break it out?
Source-stated
Yoav Shoham argues that hype about AI agents could race ahead of reality and trigger a backlash before the technology delivers real benefits.
And when expectations get out of hand, a backlash isn’t far behind.
Source-stated
The newsletter describes AI agents as software that could take on tasks such as booking flights, rescheduling meetings, and filing expenses.
The vision is exciting: Intelligent software agents that act like digital coworkers, booking your flights, rescheduling meetings, filing expenses, and talking to each other behind the scenes to get things done.
Source-stated
The newsletter reports that Google signed an agreement to purchase electricity from a fusion company’s forthcoming first power plant.
The company announced that it’s signed an agreement to purchase electricity from a fusion company’s forthcoming first power plant.
Source-stated
The newsletter reports that Google’s data-center energy use has doubled since 2020.
Google also released its latest environmental report, which shows that its energy use from data centers has doubled since 2020.
Source-stated
Casey Crownhart says Google does not break out how much of its pollution is attributable to AI.
Of course, we don’t know exactly how much of this pollution is attributable to AI because Google doesn’t break that out.
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.
AI agent
An AI system that chooses and carries out actions toward a goal, often using software tools.
Editorial provenance
Version 1. Publication ID: d28f0285-58e3-4c58-70a4-cc6ef2c5b6ce.
Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.
Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-13T05:58:45.059Z. 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: 56b651f14512e39d5dd3d0d6d38180d18290b532d3d8b9fafac36d1567f66f09. Source text hash: 09c8e5d27945668fd92e9bf3b84e47f4a998e697cc3b113f9c2f50f541466205.
Revision and source-check history
Version 1 · 2026-11-13T06:03:24.987Z
MIT Technology Review’s The Download warns against AI-agent hype and examines Google’s power demand
Dated July third, twenty twenty-five, MIT Technology Review’s The Download presents two technology concerns: expectations for and the electricity used by data centers. Yoav Shoham argues that software agents could handle tasks such as booking flights and filing expenses, but warns that hype racing ahead of reality could provoke a backlash before the technology delivers real benefits. The newsletter also reports that Google signed an agreement to buy electricity from a forthcoming fusion power plant and that its data-center energy use has doubled since 2020. Casey Crownhart’s climate-newsletter item says the announcements show big tech companies hunting for clean electricity as demand and emissions grow. It also points to a key uncertainty: Google does not break out how much of its pollution is attributable to AI. The issue links enthusiasm about what AI agents might do with the practical question of how their infrastructure will be powered; it does not establish how much AI contributes to Google’s emissions.
Original source
MIT Technology Review · Article
Source details
Original source date: Jul 3, 2025 (2025-07-03T12:10:00Z).
Date basis: Date from the publisher’s own records, through its API.
Added to Escape Meridian: 2026-11-13T06:03:24.987Z. First observed: 2026-11-01T03:47:52.811Z. Retrieved: 2026-11-01T03:49:26.809Z.
Limits
Text as retrieved on November 1, 2026; originally published July 3, 2025. The retrieved page combines newsletter items and links to fuller stories; it does not provide the full context of Shoham’s argument or further detail on Google’s electricity agreement.
Open question
How much of Google’s pollution is attributable to AI, given that the company does not break it out?
Source-stated
Yoav Shoham argues that hype about AI agents could race ahead of reality and trigger a backlash before the technology delivers real benefits.
And when expectations get out of hand, a backlash isn’t far behind.
Source-stated
The newsletter describes AI agents as software that could take on tasks such as booking flights, rescheduling meetings, and filing expenses.
The vision is exciting: Intelligent software agents that act like digital coworkers, booking your flights, rescheduling meetings, filing expenses, and talking to each other behind the scenes to get things done.
Source-stated
The newsletter reports that Google signed an agreement to purchase electricity from a fusion company’s forthcoming first power plant.
The company announced that it’s signed an agreement to purchase electricity from a fusion company’s forthcoming first power plant.
Source-stated
The newsletter reports that Google’s data-center energy use has doubled since 2020.
Google also released its latest environmental report, which shows that its energy use from data centers has doubled since 2020.
Source-stated
Casey Crownhart says Google does not break out how much of its pollution is attributable to AI.
Of course, we don’t know exactly how much of this pollution is attributable to AI because Google doesn’t break that out.
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.
AI agent
An AI system that chooses and carries out actions toward a goal, often using software tools.
Editorial provenance
Version 1. Publication ID: d28f0285-58e3-4c58-70a4-cc6ef2c5b6ce.
Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.
Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-13T05:58:45.059Z. 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: 56b651f14512e39d5dd3d0d6d38180d18290b532d3d8b9fafac36d1567f66f09. Source text hash: 09c8e5d27945668fd92e9bf3b84e47f4a998e697cc3b113f9c2f50f541466205.
uncertain · 2026-11-13T05:58:45.059Z · anthropic/claude-opus-5.5
Publication: d28f0285-58e3-4c58-70a4-cc6ef2c5b6ce
attribution — pass: The issue is one newsletter piece that comments on items in its own words, not a bare listing. The reading attributes the AI-agent warning to Yoav Shoham and the Google energy analysis to Casey Crownhart's climate-newsletter item. Google's fusion deal and environmental report are presented as the newsletter's reports, not as established facts.
—Yoav Shoham is a professor emeritus at Stanford University and cofounder of AI21 Labs.
summary — pass: The reading names the actors and states both arguments in plain language. It keeps Shoham's warning that hype could cause a backlash and Crownhart's point that Google does not break out AI's share of its pollution. The definitions list explains AI agent. The question follows directly from an uncertainty the source itself raises.
But if we’re not careful, we’re going to derail the whole idea before it has a chance to deliver real benefits.
date — pass: originalPublished is 2025-07-03T12:10Z (instant), and July 3, 2025 matches its UTC day. The Deep Dive teasers describe possibly later events, such as OpenAI's math findings, but the reading neither uses nor relies on them. Nothing is framed with hindsight. The kind is publisher-page, so no archive disclosure is required.
support — uncertain: All five claims match their quoted excerpts verbatim. The summary's closing sentence, however, says the issue links enthusiasm about AI agents with how their infrastructure will be powered. The source only places the two items side by side; Crownhart ties energy demand to AI generally, not to agents. This is unlabeled interpretation. Smallest repair: say the issue juxtaposes the two items, or label the link as Escape Meridian interpretation.
how desperately big tech companies are hunting for clean electricity to power their data centers as energy demand and emissions balloon in the age of AI
media — pass: No media was selected and no verified media candidates were supplied, so a null media selection is appropriate.
Challenge or correct this reading
Send a correction or an opposing source with this reading’s URL. Disagreement and uncertainty can remain unresolved.
An AI system that chooses and carries out actions toward a goal, often using software tools.
Definition referenceAn AI system that chooses and carries out actions toward a goal, often using software tools.
Definition reference