ІНСАЙДЕР УКРАЇНА

Robots with reasoning power are becoming a reality thanks to massive amounts of training data and breakthroughs in artificial intelligence. VOA’s Matt Dibble visits a lab where robots are learning to solve problems themselves. Cameras: Matt Dibble, Tina Trinh.
…

WASHINGTON — A bipartisan duo in the U.S. House is proposing legislation to ban the Chinese artificial intelligence app DeepSeek from federal devices, similar to the policy already in place for the popular social media platform TikTok.

Lawmakers Josh Gottheimer, a Democrat from New Jersey, and Darin LaHood, a Republican from Illinois, on Thursday introduced the “No DeepSeek on Government Devices Act,” which would ban federal employees from using the Chinese AI app on government-owned electronics. They cited the Chinese government’s ability to use the app for surveillance and misinformation as reasons to keep it away from federal networks.

“The Chinese Communist Party has made it abundantly clear that it will exploit any tool at its disposal to undermine our national security, spew harmful disinformation, and collect data on Americans,” Gottheimer said in a statement. “We simply can’t risk the CCP infiltrating the devices of our government officials and jeopardizing our national security.”

The proposal comes after the Chinese software company in January published an AI model that performed at a competitive level with models developed by American firms like OpenAI, Meta, Alphabet and others. DeepSeek purported to develop the model at a fraction of the cost of its American counterparts. The announcement raised alarm bells and prompted debates among policymakers and leading Silicon Valley financiers and technologists.

The churn over AI is coming at a moment of heightened competition between the U.S. and China in a range of areas, including technological innovation. The U.S. has levied tariffs on Chinese goods, restricted Chinese tech firms like Huawei from being used in government systems, and banned the export of state of the art microchips thought to be needed to develop the highest end AI models.

Last year, Congress and then-President Joe Biden approved a divestment of the popular social media platform TikTok from its Chinese parent company or face a ban across the U.S.; that policy is now on hold. President Donald Trump, who originally proposed a ban of the app in his first term, signed an executive order last month extending a window for a long-term solution before the legally required ban takes effect.

In 2023, Biden banned TikTok from federal-issued devices.

“The technology race with the Chinese Communist Party is not one the United States can afford to lose,” LaHood said in a statement. “This commonsense, bipartisan piece of legislation will ban the app from federal workers’ phones while closing backdoor operations the company seeks to exploit for access. It is critical that Congress safeguard Americans’ data and continue to ensure American leadership in AI.”

The bill would single out DeepSeek and any AI application developed by its parent company, the hedge fund High-Flyer, as subject to the ban. The legislation includes exceptions for national security and research purposes that would allow federal employers to study DeepSeek.

Some lawmakers wish to go further. A bill proposed last week by Senator Josh Hawley, a Republican from Missouri, would bar the import or export of any AI technology from China writ large, citing national security concerns.
…

U.S. prosecutors on Tuesday unveiled an expanded 14-count indictment accusing former Google software engineer Linwei Ding of stealing artificial intelligence trade secrets to benefit two Chinese companies he was secretly working for. 

Ding, 38, a Chinese national, was charged by a federal grand jury in San Francisco with seven counts each of economic espionage and theft of trade secrets. 

Each economic espionage charge carries a maximum 15-year prison term and $5 million fine, while each trade secrets charge carries a maximum 10-year term and $250,000 fine. 

The defendant, also known as Leon Ding, was indicted last March on four counts of theft of trade secrets. He is free on bond. His lawyers did not immediately respond to requests for comment. 

Ding’s case was coordinated through an interagency Disruptive Technology Strike Force created in 2023 by the Biden administration. 

The initiative was designed to help stop advanced technology from being acquired by countries such as China and Russia or potentially threatening national security. 

Prosecutors said Ding stole information about the hardware infrastructure and software platform that lets Google’s supercomputing data centers train large AI models. 

Some of the allegedly stolen chip blueprints were meant to give Google an edge over cloud computing rivals Amazon and Microsoft, which design their own, and reduce Google’s reliance on chips from Nvidia. 

Prosecutors said Ding joined Google in May 2019 and began his thefts three years later when he was being courted to join an early-stage Chinese technology company. 

Ding allegedly uploaded more than 1,000 confidential files by May 2023 and later circulated a PowerPoint presentation to employees of a China startup he founded, saying that country’s policies encouraged development of a domestic AI industry. 

Google was not charged and has said it cooperated with law enforcement. 

According to court records describing a December 18 hearing, prosecutors and defense lawyers discussed a “potential resolution” to Ding’s case, “but anticipate the matter proceeding to trial.” 

The case is U.S. v. Ding, U.S. District Court, Northern District of California, No. 24-cr-00141. 
…

PARIS — France hosts top tech players next week at an artificial intelligence summit meant as a “wake-up call” for Europe as it struggles with AI challenges from the United States and China.

Players from across the sector and representatives from 80 nations will gather in the French capital on February 10 and 11 in the sumptuous Grand Palais, built for the 1900 Universal Exhibition.

In the run-up, President Emmanuel Macron will on Feb. 4 visit research centers applying AI to science and health, before hosting scientists and Nobel Prize winners at his Elysee Palace residence on Wednesday.

A wider science conference will be held at the Polytechnique engineering school on Thursday and Friday.

“The summit comes at exactly the right time for this wake-up call for France and Europe, and to show we are in position” to take advantage of the technology, an official in Macron’s office told reporters.

In recent weeks, Washington’s announcement of $500 billion in investment to build up AI infrastructure and the release of a frugal but powerful generative AI model by Chinese firm DeepSeek have focused minds in Europe.

France must “not let this revolution pass it by,” Macron’s office said.

Attendees at the summit will include Sam Altman, head of OpenAI — the firm that brought generative models to public consciousness in 2022 with the launch of ChatGPT.

Google boss Sundar Pichai and Nobel Prize winner Demis Hassabis, who leads the company’s DeepMind AI research unit, will also come, alongside Arthur Mensch, founder of French AI developer Mistral.

The Elysee has said there are “talks” on hosting DeepSeek founder Liang Wenfeng, and has yet to clarify whether X owner Elon Musk — who has his own generative initiative, xAI — has accepted an invitation.

Nor is it clear who will attend from the United States and China, with the French presidency saying only “very high level” representatives will come.

Confirmed guests from Europe include European Commission chief Ursula von der Leyen and German Chancellor Olaf Scholz.

‘Stoke confidence’

The tone of the AI summit will be “neither catastrophizing, nor naive,” Macron’s AI envoy Anne Bouverot told AFP.

Hosting the conference is also an opportunity for Paris to show off its own AI ecosystem, which numbers around 750 companies.

Macron’s office has said the summit would see the announcement of “massive” investments along the lines of his annual “Choose France” business conference, at which $15.4 billion of inward investment were pledged in 2024.

Beyond the economic opportunities, AI’s impact on culture including artistic creativity and news production will be discussed in a side-event over the weekend.

Debates open to the public, such as that one, are aimed at showing off “positive use cases for AI” to “stoke confidence and speed up adoption” of the technology, said France’s digital minister Clara Chappaz.

For now, the French public is skeptical of AI, with 79 percent of respondents telling pollsters Ifop they were “concerned” about the technology in a recent survey.

More ‘inclusive’ AI?

Paris says it also hopes the summit can help kick off its vision of a more ethical and accessible and less resource-intensive AI.

At present, “the AI under development is pushed by a few large players from a few countries,” Bouverot said, whereas France wants “to promote more inclusive development.”

Indian Prime Minister Narendra Modi has been invited to co-host the Paris summit, in a push to bring governments on board.

One of the summit’s aims is the establishment of a public-interest foundation for which Paris aims to raise $2.5 billion over five years.

The effort would be “a public-private partnership between various governments, businesses and philanthropic foundations from different countries,” Macron’s office said.

Paris hopes at the summit to chart different efforts at AI governance around the world and gather commitments for environmentally sustainable AI — although no binding mechanism is planned for now.

“There are lots of big principles emerging around responsible, trustworthy AI, but it’s not clear or easy to implement for the engineers in technical terms,” said Laure de Roucy-Rochegonde, director of the geopolitical technology center at the French Institute for International Relations.
…