Commentary: Is DeepSeek triggering a Sputnik moment for the American AI industry?

By:
On:
In: Asia

BLIND SPOTS EXPOSED

Although there is some disagreement over whether DeepSeek has told the entire history, this show has exposed “groupthink” in the British AI industry. Its deafness to solution, cheaper, more encouraging approaches, combined with enthusiasm, is exactly what Simon Johnson and I predicted in Power and Progress, which we wrote just before the generative-AI time began. &nbsp,

The question is then whether the US economy has any other, even more risky blind spots. For instance, are the leading US tech firms missing an opportunity to get their designs in a more “pro-human way”? I’m sure the answer may be well, but only time will tell.

Then there is the issue of whether China is outperforming the US. If so, does this mean that authoritarian, top-down structures ( what James A Robinson and I have called “extractive institutions” ) can match or even outperform bottom-up arrangements in driving innovation?

My discrimination is to believe that top-down power hampers development, as Robinson and I argued in Why Governments Fail. Although DeepSeek’s success makes it seem unlikely to dispute this assertion, it is far from convincing evidence that innovation can be as potent or resilient under inclusive institutions. &nbsp,

After all, DeepSeek is building on decades of advancements in the US and some of Europe. In the US, all of its fundamental techniques were used as models. Mixture-of-experts designs and support understanding were developed in scientific research organizations decades ago, and it was US Big Tech firms that introduced converter models, chain-of-thought argument, and evaporation.

What DeepSeek has accomplished is exhibit its engineering success: utilizing the same techniques more efficiently than US businesses. Will Chinese companies and research institutes be able to develop innovative products, techniques, and strategies that will change the game in the future.

In addition, DeepSeek appears to be different from most other Chinese Artificial companies, which typically develop technology for the state or receive funding from the government. Does the company’s creativity and vitality continue now that it is in the light if it ( which was spun out of a hedge fund ) was operating under the radar? &nbsp,

Whatever happens, one company’s success cannot be taken as compelling evidence that China can defeat more open cultures at technology.

Continue Reading

Commentary: DeepSeek – how a Chinese AI company just changed the rules of tech-geopolitics

By:
On:
In: Asia

Washington’s steps were part of its so-called” small-yard-high-fence” method, which aims to control who gets exposure to corporate American technology and who gets locked out of the garden. &nbsp,

But perhaps the most significant result was that, from a techno-nationalist view, Washington actually sorted the world’s governments into three different groups. By design, this activity was intended to exacerbate the division of the global tech landscape into distinct subsets.

America’s former allies and surveillance partners, who all have generally unrestricted access to the yard, are included in the second class. This includes Canada, Australia, the UK and Western Europe, Germany, France, Japan and South Korea.

The next group consists of a large chunk of “middle-countries” including those in Eastern Europe, India, Saudi Arabia, the UAE, and all of Southeast Asia.

These countries face higher limits, and, by architecture, Washington may push them to pick sides between the US and China when it comes to trading and partnering around corporate systems. This would include the creating of AI systems, information areas, energy infrastructure and additional power-multiplier systems.

The third class, consisting of China, Russia, Iran and North Korea, faces the most constraints, and is basically banned from it move from Group 1 and Group 2. Those who pass restricted technology to these will be targeted and punished.

However, they may also avoid having to pick sides if public and private entities in any of the three country groups can use open-sourced conceptual AI ( and finally artificial general intelligence ) to achieve their individual goals and lessen their dependence on the two international tech superpowers. This would change the present-day techno-nationalist calculus and give more agency to (especially ) middle-tier countries.

Continue Reading