# MIT Technology Review weighs competing visions of AI’s next chapter [Canonical reading](https://escapemeridian.com/readings/mit-technology-review-weighs-competing-visions-of-ai-s-next-1e0a3067) MIT Technology Review published this final edition on December 8, 2025, as a conversation between senior AI editor Will Douglas Heaven and Financial Times correspondent Tim Bradshaw. Heaven’s answer to what the world will look like in 2030 is “same but different”: rapid advances in AI, he argues, do not translate straight away into economic or social change because adoption and acceptance move at human speed. He expects applications of existing models to matter more as progress in core technology slows, while other approaches, including reinforcement learning and world models, may gain attention. Bradshaw expects a more unequal outcome. He argues that the costs of running AI services could price people and businesses out even as those able to pay gain productivity benefits. He predicts that AI app developers will disappear after a funding shakeout, and that robotaxis and home robots could remain luxuries. Heaven agrees that costs could widen gaps, but remains skeptical that the promised revolution will arrive by 2030; casual use, he says, is not the same as lasting technological change. The exchange ends with Heaven joking that Bradshaw can send him laundry by robotaxi if he gets a domestic robot—“As if I could afford one.” How much AI’s costs will limit access, and whether cheaper approaches or models can change that, remains unresolved. 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: Dec 8, 2025 (2025-12-08T16:30:00Z). Date basis: Date from the publisher’s own records, through its API. Added to Escape Meridian: 2026-11-09T18:28:27.164Z. First observed: 2026-11-02T00:19:27.815Z. Retrieved: 2026-11-02T00:21:23.802Z. - [The State of AI: A vision of the world in 2030](https://www.technologyreview.com/2025/12/08/1128922/the-state-of-ai-a-vision-of-the-world-in-2030/) ## Limits Text as retrieved on November 2, 2026; originally published December 8, 2025. The piece presents a debate and forecasts by Heaven and Bradshaw; it does not establish how their predictions will turn out. ## Open question Can cheaper models or more efficient computing make useful AI services broadly affordable, or will the costs keep access concentrated among those able to pay? ## Source-stated Heaven argues that rapid progress in AI does not immediately produce broad economic or social change, because adoption and acceptance take time. > rapid technological advances do not translate to economic or societal ones straight away - [Supporting source](https://www.technologyreview.com/2025/12/08/1128922/the-state-of-ai-a-vision-of-the-world-in-2030/) ## Source-stated Heaven expects applications of existing models to become a key area of competition as progress in core technology slows. > As the rate of advance in the core technology slows down, applications of that tech will become the main differentiator between AI firms. - [Supporting source](https://www.technologyreview.com/2025/12/08/1128922/the-state-of-ai-a-vision-of-the-world-in-2030/) ## Source-stated Bradshaw predicts that AI costs could leave people and businesses unable to afford services while those able to pay gain productivity benefits. > Those able to pay will reap the productivity benefits, scooping up the excess computing power as others are priced out of the market. - [Supporting source](https://www.technologyreview.com/2025/12/08/1128922/the-state-of-ai-a-vision-of-the-world-in-2030/) ## Source-stated Bradshaw expects robotaxis and humanoid robots to become common, but predicts that their costs could make them luxuries for the well-off, at least in the near term. > the compute cost involved in making them useful and ubiquitous seems destined to turn them into luxuries for the well-off, at least in the near term. - [Supporting source](https://www.technologyreview.com/2025/12/08/1128922/the-state-of-ai-a-vision-of-the-world-in-2030/) ## Source-stated Heaven remains skeptical that the revolution promised by many insiders will arrive by 2030. > I still remain skeptical that we will see anything like the revolution that many insiders promise (and investors pray for) by 2030. - [Supporting source](https://www.technologyreview.com/2025/12/08/1128922/the-state-of-ai-a-vision-of-the-world-in-2030/) ## 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. ## Reinforcement learning Training by trial and error, in which a system is rewarded for the outcomes its designers want. - [Definition reference](https://developers.google.com/machine-learning/glossary#reinforcement-learning-rl) ## World model An AI system's internal model of how an environment behaves, used to predict what happens next or to plan. - [Definition reference](https://arxiv.org/abs/1803.10122) ## The World The World is a public radio news program publishing international reporting on politics, society, labor and technology. - [Explore this context](/entities/the-world) - [Reference](https://theworld.org/) ## Editorial provenance Version 1. Publication ID: 74dac332-f154-7070-66c2-1c8901b28f06. Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5. Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-09T18:23:45.562Z. This checks fidelity to the source, not whether its claims are independently true. Editorial selection: accept; reasons: bridge\_admitted\_by\_model. Observation: 61f4ccb8197cad1babda1ec4117dc8b93234c4f289ee15619f43c42d5e1cc268. Source text hash: 43a52012722877a05b2f7b85b9d4403fc66224477212c4d70c0b30ba061b014e. ## Revision and source-check history ### Version 1 — 2026-11-09T18:28:27.164Z MIT Technology Review published this final edition on December 8, 2025, as a conversation between senior AI editor Will Douglas Heaven and Financial Times correspondent Tim Bradshaw. Heaven’s answer to what the world will look like in 2030 is “same but different”: rapid advances in AI, he argues, do not translate straight away into economic or social change because adoption and acceptance move at human speed. He expects applications of existing models to matter more as progress in core technology slows, while other approaches, including reinforcement learning and world models, may gain attention. Bradshaw expects a more unequal outcome. He argues that the costs of running AI services could price people and businesses out even as those able to pay gain productivity benefits. He predicts that AI app developers will disappear after a funding shakeout, and that robotaxis and home robots could remain luxuries. Heaven agrees that costs could widen gaps, but remains skeptical that the promised revolution will arrive by 2030; casual use, he says, is not the same as lasting technological change. The exchange ends with Heaven joking that Bradshaw can send him laundry by robotaxi if he gets a domestic robot—“As if I could afford one.” How much AI’s costs will limit access, and whether cheaper approaches or models can change that, remains unresolved. ## Original source MIT Technology Review · Article Original source date: Dec 8, 2025 (2025-12-08T16:30:00Z). Date basis: Date from the publisher’s own records, through its API. Added to Escape Meridian: 2026-11-09T18:28:27.164Z. First observed: 2026-11-02T00:19:27.815Z. Retrieved: 2026-11-02T00:21:23.802Z. - [The State of AI: A vision of the world in 2030](https://www.technologyreview.com/2025/12/08/1128922/the-state-of-ai-a-vision-of-the-world-in-2030/) ## Limits Text as retrieved on November 2, 2026; originally published December 8, 2025. The piece presents a debate and forecasts by Heaven and Bradshaw; it does not establish how their predictions will turn out. ## Open question Can cheaper models or more efficient computing make useful AI services broadly affordable, or will the costs keep access concentrated among those able to pay? ## Source-stated Heaven argues that rapid progress in AI does not immediately produce broad economic or social change, because adoption and acceptance take time. > rapid technological advances do not translate to economic or societal ones straight away - [Supporting source](https://www.technologyreview.com/2025/12/08/1128922/the-state-of-ai-a-vision-of-the-world-in-2030/) ## Source-stated Heaven expects applications of existing models to become a key area of competition as progress in core technology slows. > As the rate of advance in the core technology slows down, applications of that tech will become the main differentiator between AI firms. - [Supporting source](https://www.technologyreview.com/2025/12/08/1128922/the-state-of-ai-a-vision-of-the-world-in-2030/) ## Source-stated Bradshaw predicts that AI costs could leave people and businesses unable to afford services while those able to pay gain productivity benefits. > Those able to pay will reap the productivity benefits, scooping up the excess computing power as others are priced out of the market. - [Supporting source](https://www.technologyreview.com/2025/12/08/1128922/the-state-of-ai-a-vision-of-the-world-in-2030/) ## Source-stated Bradshaw expects robotaxis and humanoid robots to become common, but predicts that their costs could make them luxuries for the well-off, at least in the near term. > the compute cost involved in making them useful and ubiquitous seems destined to turn them into luxuries for the well-off, at least in the near term. - [Supporting source](https://www.technologyreview.com/2025/12/08/1128922/the-state-of-ai-a-vision-of-the-world-in-2030/) ## Source-stated Heaven remains skeptical that the revolution promised by many insiders will arrive by 2030. > I still remain skeptical that we will see anything like the revolution that many insiders promise (and investors pray for) by 2030. - [Supporting source](https://www.technologyreview.com/2025/12/08/1128922/the-state-of-ai-a-vision-of-the-world-in-2030/) ## 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. ## Reinforcement learning Training by trial and error, in which a system is rewarded for the outcomes its designers want. - [Definition reference](https://developers.google.com/machine-learning/glossary#reinforcement-learning-rl) ## World model An AI system's internal model of how an environment behaves, used to predict what happens next or to plan. - [Definition reference](https://arxiv.org/abs/1803.10122) ## The World The World is a public radio news program publishing international reporting on politics, society, labor and technology. - [Explore this context](/entities/the-world) - [Reference](https://theworld.org/) ## Editorial provenance Version 1. Publication ID: 74dac332-f154-7070-66c2-1c8901b28f06. Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5. Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-09T18:23:45.562Z. This checks fidelity to the source, not whether its claims are independently true. Editorial selection: accept; reasons: bridge\_admitted\_by\_model. Observation: 61f4ccb8197cad1babda1ec4117dc8b93234c4f289ee15619f43c42d5e1cc268. Source text hash: 43a52012722877a05b2f7b85b9d4403fc66224477212c4d70c0b30ba061b014e. ### uncertain — 2026-11-09T18:23:45.562Z — anthropic/claude-opus-5.5 Publication: 74dac332-f154-7070-66c2-1c8901b28f06 attribution — pass: The reading identifies a conversation between MIT Technology Review's senior AI editor Will Douglas Heaven and FT correspondent Tim Bradshaw. It assigns each view to the correct speaker. The title credits only MIT Technology Review, though the series is an FT/MIT Technology Review collaboration. The summary names both writers, so this does not misattribute anything. > MIT Technology Review’s senior AI editor Will Douglas Heaven talks with Tim Bradshaw, FT global tech correspondent summary — uncertain: The reading is clear and the terms are defined, but it softens Bradshaw's conviction. His confident predictions become hedged: costs 'could price people out' and robots 'could remain luxuries', where he says 'will' and 'seems destined'. It also drops his central claim that the AI bubble will inevitably burst before 2030. And 'AI app developers will disappear' omits his 'swathes of'. Smallest repair: state that Bradshaw calls a bubble burst inevitable, keep his 'will'/'destined' wording, and restore 'swathes of'. > It seems inevitable that the AI bubble will burst sometime before the end of the decade. date — pass: The summary gives December 8, 2025, matching the UTC day of originalPublished. The forecasts are framed as the authors' predictions, not outcomes. The later Deep Dive sidebar items, such as the AlphaGo ten-year piece, are not used. This is a publisher page, so no archive disclosure is needed. support — pass: Each claim matches its verbatim excerpt and is attributed to the right speaker. The 'adoption and acceptance' rationale comes from the Normal Technology view that Heaven explicitly endorses. The closing question follows from Bradshaw's remarks on an efficiency breakthrough and from the open-weight counter-argument. The reading does not invent benefits or costs. > Perhaps some breakthrough in computational efficiency will avert this fate. media — pass: No media was selected, and no verified candidates were supplied, so nothing needs checking. ## 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/1e0a30676229a7f69227ce88e80c3c3fe8580eaa46023293a014d7f6ec37c507/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-weighs-competing-visions-of-ai-s-next-1e0a3067.json)