Dirk Mattig’s argument is that people may gain more by learning to use today’s AI tools than by fixating on the next model leap. That is a personal view about how to make practical progress—not evidence that AI capability development has slowed. The debate also includes a different concern: Anthropic CEO Dario Amodei has proposed pacing frontier development so safety work and independent evaluation can keep up, while explicitly saying this need not halt technical progress.
What Mattig means by “slowing down”
In his September 15, 2026 essay on DEV Community, Dirk Mattig questions whether safety is the only reason people are discussing a slower pace for AI. He writes, “I am not convinced that safety concerns are the only reason behind this AI slowdown debate.” He does not claim to know how much legal, financial, or commercial considerations shape the debate.
As an Amazon Associate I earn from qualifying purchases.
His practical point is about attention: current AI tools already offer plenty for people to learn and apply. “An industry constantly being distracted and interrupted by the next big leap forward does not necessarily deliver productivity gains,” he argues. In his view, pursuing novelty is not the same as putting existing capabilities to useful work. He closes with a personal plea: “I don’t know about you, but I could use a bit of a breather.”
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →That is an argument for catching up with tools already available, not a measured finding that users or businesses will become more productive by slowing development. Mattig’s essay offers a perspective, not an empirical test.
#1 Best Overall
How Mattig’s view differs from Amodei’s pacing proposal
The two arguments focus on different kinds of progress. Mattig is concerned with whether people and organizations are making practical use of existing tools. Amodei’s proposal addresses how quickly frontier AI systems should be developed, with the aim of giving safety work and external evaluation time to keep pace.
| Position | Goal | Proposed mechanism | Evidence status |
|---|---|---|---|
| Dirk Mattig’s essay | Encourage people to apply current AI tools rather than treat each new leap as the main source of progress. | Individual and organizational learning and adoption. | Personal argument; it does not measure productivity or frontier capability progress. |
| Dario Amodei’s essay | Give safety work and independent evaluation adequate time as frontier capabilities develop. | Pacing, embedded third-party evaluators, and coordination at democratic and global levels. | Policy proposal, not evidence that capability progress has slowed. |
Amodei states: “To be clear, pacing does not mean halting model training or technical progress, but ensuring companies take adequate time to align and safeguard their models, and for third party evaluators to confirm this.” His proposal is therefore about the conditions and oversight around development, not stopping it.
Rank #2
What business adoption data can—and cannot—tell us
There is a separate question from whether frontier models are improving: are organizations getting financial returns from AI? PwC’s 29th Global CEO Survey, published January 19, 2026, surveyed 4,454 CEOs in 95 countries and territories. The survey found that 56% of respondents said their companies had realized neither higher revenue nor lower costs from AI, while 12% reported both. PwC summarized the finding this way: “Most CEOs say their companies aren’t yet seeing a financial return from investments in AI.”
Free tools Windows power users keep installed
One-click scans. No signup required.
These are CEOs’ reported business outcomes, based on a survey conducted September 30–November 10, 2025. They do not measure AI capabilities, prove that AI cannot produce value, or establish whether model progress has slowed. They do illustrate why Mattig’s distinction matters: the arrival of a more capable tool does not automatically mean an organization has translated it into revenue or lower costs.
Does the slowdown debate prove AI progress has slowed?
No. A public debate about pacing, or a proposal to pace development, does not establish that frontier-model progress has actually slowed. The sources discussed here do not provide a time series or an independent measurement that settles that technical question.
Legal claims also need to be kept separate from evidence of a technical slowdown. An Associated Press report published by OPB on September 20, 2026, said four paid AI subscribers filed a proposed class action in the Northern District of California on September 18. The complaint alleges that Anthropic, OpenAI, SpaceXAI, and Google coordinated to slow development. Those claims are allegations; the report does not describe a court finding that such an agreement existed.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What to take from Mattig’s argument
Mattig’s essay is most useful as a reminder to assess progress at the level where people encounter AI: the work they can do with available tools. His view invites a practical question—whether time spent learning and applying current capabilities may be more useful to ordinary users than waiting for another major leap. It does not answer that question for every person or organization, and it does not establish that development should pause.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Amodei’s proposal addresses a different concern: whether safety measures and independent evaluation can keep up with frontier development. These positions are not competing test results. One emphasizes putting current tools to work; the other emphasizes oversight as capabilities develop. Neither, on its own, shows that AI progress has slowed.
Quick Recap
Best Value
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
