I recently realized a fundamental difference between the style of regulation employed by the USA, the
European Union and China. While the EU tries their best at predicting the future and preemptively regulate
industries, China drops their regulations when problems arise (often a bit too late). The US on the
other hand, does what their oligarchs lobbyists tell them to do.
Note that this post is by no means an in depth research piece, just an observation I made with some examples. The two industries I will use to support my thesis are AI and electric vehicles.
In 2021, the EU released 'Fit for 55', the climate law codifying its goals of reducing EU emmisions by at
least 55% by 2030 [1]. One of the main pieces of the regulation was requiring 100% CO2 reduction for new
cars by 2035. Effectively banning internal combustion engine (ICE) cars.
2021 being one of the high points of the EV boom, this goal was ambitious, yet reachable, the EU figured.
The issue was that there was no internal infrastructure for building and supporting that many EVs. The
batteries come from China, there were not enough charging stations electric cars lose value much quicker
than their ICE counterparts, inhibiting widespread adoption (when not subsidised, that is). These issues
were (or should have been) known at the time, and can be labeled as a misjudgment, assuming the plan was to
fix these issues before 2035.
Other issues, such as the AI boom increasing the price of electricity [2] and deglobalisation & tensions
with china making it more expesive to obtain high quality batteries were less easy to predict. The point
being, that predicting what the future will hold and writing regulation based on this prediction and your
desires, is doomed to fail.
As of december 2025, the EU has walked back some of their regulation goals, such as the 100% ICE ban by
2035 [3].
Now, in 2025 I see the European Union making the same mistake again, but with the AI Act [4]. The act is
filled with semi-vague guidelines an AI system should- and should not adhere to, based on the bureaucrats
ideas of how AI works (these people know nothing about technology, let alone the intricacies of machine
learning and what models can- and can not do, but that is a different story). These regulations not only
inhibit local companies from developing world class AI systems, it makes existing companies consider leaving
the EU altogether.
Like with the ICE vehicle regulations, these regualtions are introduced during the growth of the industry,
with the finish line not in sight. At this point, anything can change and new breakthroughs will lead to
completely different outcomes which would have been impossible to predict a year earlier, let alone 10 years
like in Fit for 55.
The story of EVs in the US is slightly different. There was less EV talk from the government and more from the companies themselves. After seeing what EV hype did to $TSLA, every company started making promises regarding their EV options ten years from then (most of which have been walked back now) [5, 6]. Though the government implemented some regulation targets, aggresive lobbying (and perhaps some generous donations) have lead to the easing of regulations [7].
Like with vehicles, the AI lobby in America tries with all their might and money to stop regulation from being pushed through congress. In a single year, Meta, Alphabet, Microsoft and others have spent tens of millions on lobbying their lobbying efforts [8] and hundreds of millions have been spent on super PACs [9].
While buying (a lack of) regulation might seem like a positive thing for business growth, the fact that your government can be bought at all is probably an issue which will eventually lead to internal rot and distrust of the system, something we are seeing unravel at this very moment in the US, caused by the fact that corruption is happening so blatent and shamelessly that it is hard to ignore, even for their own team [10, 11, 12]. Even ignoring that, the flip-flopping on regulations with every new administration (especcialy these days with heightened partisanship) brings uncertainty, which inhibits growth.
While China is probably even more bureaucratic than the EU (or any other government, for that matter), it is implemented slightly differently, meaning it avoids some of the problems that the EU has, while introducing other issues. The way China likes to work with these new industries is set some high level vague goal, and allow for a wild west-like period of choas, ruthless competition and fraud. After this period, the winners are identified a strong (but often thought out) regulations are introduced to clean up the mess. Some examples of industries where this strategy is very clear are the Chinese P2P lending crisis in 2015 [13], the property sector regulations [14, 15] and the online tutoring crackdowns [16].
As for EVs, in 2020 China did not immediately set a 100% reduction goal. They started with a pragmatic target of 50% [17]. China then started subsidising the industry as well as encouraging competition. Some postive results of this being the production of great cars for a low price, as well as much innovation on stronger and longer lasting battery technology. Negative results include overproduction and EV graveyards filled with unsold cars [18]. As a response to this, China introduced legislation blocking manufactures from opening new factories if their existing are running below a certain capacity [19].
The same is currently unfolding in the AI landscape in China. Strong competition and little regulation leads to great (open source!) models, coming close to, or in many cases outcompeting their Western equivalents. As far as I know, currently there have not been big AI regulations introduced as a reaction to the industry becoming too chaotic for the CCP's liking. I predict however that in 2026 some large changes will be happening in China's AI sector as a response to unexpected chaos or some big scandal.
While none of these strategies are perfect by any means, I think there is a lot to learn from each other. The EU should probably allow for some natural competition and innovation to take place before rushing to make everything illegal. US leadership should probably be a bit less money hungry, and both bring in more engineers in the bureaucracy [20, 21]. China on the other hand should probably strive to make their regulatory landings a bit softer to prevent more hundreds of billions of dollars from being lost after the regulatory shocks [16].
[1]: https://www.consilium.europa.eu/en/policies/fit-for-55/
[2]: https://www.iea.org/news/ai-is-set-to-drive-surging-electricity-demand-from-data-centres-while-offering-the-potential-to-transform-how-the-energy-sector-works
[3]: https://ec.europa.eu/commission/presscorner/detail/en/ip_25_3051
[4]: https://artificialintelligenceact.eu/#
[5]: https://www.cnbc.com/2021/01/28/general-motors-plans-to-exclusively-offer-electric-vehicles-by-2035.html
[6]: https://www.npr.org/2025/12/15/nx-s1-5645147/ford-discontinues-all-electric-f-150-lightning
[7]: https://www.movinon.eu/en/nos-actualites/epa-final-pollution-standards-vehicles-2027-2032/
[8]: https://issueone.org/articles/as-washington-debates-major-tech-and-ai-policy-changes-big-techs-lobbying-is-relentless/
[9]: https://medium.com/@nemko_digital/how-big-tech-lobbying-halted-us-ai-regulation-in-2025-1306dddddfc4
[10]: https://www.reuters.com/world/us/trump-draws-global-crypto-investors-with-148-million-meme-coin-dinner-2025-05-22/
[11]: https://abcnews.go.com/Politics/us-accepts-unconditional-donation-qatari-jet-cost-retrofitting/story?id=124150583
[12]: https://www.reuters.com/business/finance/kushners-affinitys-assets-jump-48-billion-after-gulf-cash-injection-2025-03-28/
[13]: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3869264
[14]: https://english.ckgsb.edu.cn/knowledge/professor_analysis/series-chinas-real-estate-problem-1-the-three-red-lines
[15]: https://www.researchgate.net/publication/373896412_The_Impact_of_Three_Red_Lines_Policy_on_Chinas_Real_Estate_Industry
[16]: https://www.classcentral.com/report/china-regulatory-tsunami
[17]: https://www.reuters.com/technology/chinas-nev-sales-account-20-new-car-sales-by-2025-50-by-2035-2020-10-27/
[18]: https://www.bloomberg.com/features/2023-china-ev-graveyards/
[19]: https://www.bloomberg.com/news/articles/2021-09-08/china-eyeing-ways-to-better-target-resources-for-ev-production
[20]: https://freakonomics.com/podcast/china-is-run-by-engineers-america-is-run-by-lawyers/
[21]: https://danwang.co/breakneck/