The Download: how the World Cup ball will fly and OpenAI’s “super app”
MIT Technology Review’s newsletter leads with research on the new World Cup ball’s flight, then rounds up major AI stories, including OpenAI’s super-app plans.
Intelligence analysis by GPT-5.4 Mini
This Download edition opens with wind-tunnel research suggesting Adidas’s new World Cup ball may travel less far but fly more predictably. It then pivots to a fast-moving AI roundup, led by reports that OpenAI wants to turn ChatGPT into a super app before its IPO.
This newsletter says a new World Cup ball might not fly as far, but it should fly in a steadier way. It also says OpenAI wants to turn ChatGPT into a bigger all-in-one app, like a Swiss Army knife for computer jobs.
Analysis
Lead story
The edition starts with a sports-science angle: researchers used wind-tunnel experiments to study Adidas’s new Trionda ball for this year’s FIFA World Cup. Their finding is that long-distance kicks may not go as far as in previous tournaments, but the tradeoff is a more predictable flight path. The piece frames that as part of the long tradition of World Cup balls being redesigned for each tournament, while noting that this year’s event is unusually big and spread across three host countries.
The AI roundup
The rest of the newsletter is a compact tour of current tech news, with AI at the center. The biggest item says OpenAI plans to turn ChatGPT into a “super app” before its IPO, combining coding tools and AI agents. Another item says the US president wants the government to take stakes in AI companies, while Google has reportedly agreed to pay SpaceX for large-scale AI computing power. The newsletter also highlights a claim that AI could make everyday life more expensive by driving up inflation through resource demand.
Broader themes
Other links in the roundup point to Europe’s effort to move away from US Big Tech, ICE’s planned facial-recognition app for local police, concerns about AI “recursive self-improvement,” and the continued tension between new biotechnologies and safety gaps. The issue ends with a lighter “One More Thing” about how machine learning is helping historians analyze older records, while also warning that the same tools can introduce bias or falsehoods into historical work.
Taken together, the edition frames AI as a shift from chatbots toward agents, infrastructure, policy fights, and new risks that reach well beyond software demos.
Key points
- Researchers say Adidas’s new Trionda World Cup ball may travel less far on long kicks, but it should fly more predictably.
- The newsletter says OpenAI wants to turn ChatGPT into a super app before its IPO, combining coding tools and AI agents.
- The roundup highlights worries that AI demand could raise costs through energy and other resource pressures.
- Other items touch on government stakes in AI companies, facial recognition for immigration checks, and Europe’s move away from US Big Tech.
- The issue also notes that machine learning is helping historians study the past, while introducing new risks of bias and falsehoods.
If OpenAI’s plan works, ChatGPT could become a more useful place for coding and agent-based tasks instead of just chatting. The broader AI roundup also suggests the technology is pushing into real infrastructure and research, which could make tools more capable and practical.
The same roundup points to serious pressure points: higher energy use, rising costs, government involvement, and surveillance concerns. If AI keeps expanding without clear guardrails, the result could be more expensive services, more public distrust, and more policy fights.
