TikTok Faces Tough Questions from Court over Challenge to US Law

A view of the E. Barrett Prettyman US Court House in Washington, DC on September 16, 2024. (AFP)
A view of the E. Barrett Prettyman US Court House in Washington, DC on September 16, 2024. (AFP)
TT

TikTok Faces Tough Questions from Court over Challenge to US Law

A view of the E. Barrett Prettyman US Court House in Washington, DC on September 16, 2024. (AFP)
A view of the E. Barrett Prettyman US Court House in Washington, DC on September 16, 2024. (AFP)

A lawyer for TikTok and Chinese parent company ByteDance faced tough questions on Monday as a US appeals court heard arguments in their lawsuit seeking to block a law that could ban the short video app used by 170 million Americans as soon as Jan. 19.

A three-judge panel of the US Court of Appeals for the District of Columbia began hearing arguments in the suit filed by TikTok and ByteDance in May seeking an injunction barring the law from taking effect.

The judges questioned TikTok's outside lawyer Andrew Pincus, who argued that the US government had not demonstrated that TikTok poses national security risks and that the law violates the US Constitution on a number of grounds including running afoul of First Amendment free speech protections.

"The law before this court is unprecedented, and its effect would be staggering," Pincus told the judges, saying "for the first time in history, Congress has expressly targeted a specific US speaker, banning its speech and the speech of 170 million Americans."

The law gives ByteDance until Jan. 19 to sell or divest TikTok's US assets or face a ban in the United States. Driven by worries among American lawmakers that China could access data on Americans or spy on them with the app, the US Congress passed the measure with overwhelming support and President Joe Biden signed it into law in April.

The law prohibits app stores like Apple and Alphabet's Google from offering TikTok and bars internet hosting services from supporting TikTok unless ByteDance divests TikTok by the deadline. Under the law, Biden could extend the deadline by three months if he certifies ByteDance is making significant progress toward a sale.

Judges Sri Srinivasan, Neomi Rao and Douglas Ginsburg were hearing the arguments.

The case is playing out during the final weeks of the US presidential campaign. Republican presidential candidate Donald Trump and Vice President Kamala Harris, his Democratic rival, are active on TikTok, seeking to court younger voters.

The Justice Department has said TikTok, under Chinese ownership, poses a serious national security threat because of its access to vast amounts of personal data on Americans, asserting China can covertly manipulate information that Americans consume via TikTok.

ByteDance has said divestiture is "not possible technologically, commercially or legally."

TikTok and the Justice Department have asked for a ruling by Dec. 6, which could allow the US Supreme Court to consider any appeal before a ban takes effect.

The White House has said it wants to see Chinese-based ownership ended on national security grounds, but not a ban on TikTok. Trump, who unsuccessfully tried to ban TikTok in 2020, has said if elected in November he would not allow TikTok to be banned.



AI Experts Ready ‘Humanity’s Last Exam’ to Stump Powerful Tech

Figurines with computers and smartphones are seen in front of the words "Artificial Intelligence AI" in this illustration taken, February 19, 2024. (Reuters)
Figurines with computers and smartphones are seen in front of the words "Artificial Intelligence AI" in this illustration taken, February 19, 2024. (Reuters)
TT

AI Experts Ready ‘Humanity’s Last Exam’ to Stump Powerful Tech

Figurines with computers and smartphones are seen in front of the words "Artificial Intelligence AI" in this illustration taken, February 19, 2024. (Reuters)
Figurines with computers and smartphones are seen in front of the words "Artificial Intelligence AI" in this illustration taken, February 19, 2024. (Reuters)

A team of technology experts issued a global call on Monday seeking the toughest questions to pose to artificial intelligence systems, which increasingly have handled popular benchmark tests like child's play.

Dubbed "Humanity's Last Exam," the project seeks to determine when expert-level AI has arrived. It aims to stay relevant even as capabilities advance in future years, according to the organizers, a non-profit called the Center for AI Safety (CAIS) and the startup Scale AI.

The call comes days after the maker of ChatGPT previewed a new model, known as OpenAI o1, which "destroyed the most popular reasoning benchmarks," said Dan Hendrycks, executive director of CAIS and an advisor to Elon Musk's xAI startup.

Hendrycks co-authored two 2021 papers that proposed tests of AI systems that are now widely used, one quizzing them on undergraduate-level knowledge of topics like US history, the other probing models' ability to reason through competition-level math. The undergraduate-style test has more downloads from the online AI hub Hugging Face than any such dataset.

At the time of those papers, AI was giving almost random answers to questions on the exams. "They're now crushed," Hendrycks told Reuters.

As one example, the Claude models from the AI lab Anthropic have gone from scoring about 77% on the undergraduate-level test in 2023, to nearly 89% a year later, according to a prominent capabilities leaderboard.

These common benchmarks have less meaning as a result.

AI has appeared to score poorly on lesser-used tests involving plan formulation and visual pattern-recognition puzzles, according to Stanford University’s AI Index Report from April. OpenAI o1 scored around 21% on one version of the pattern-recognition ARC-AGI test, for instance, the ARC organizers said on Friday.

Some AI researchers argue that results like this show planning and abstract reasoning to be better measures of intelligence, though Hendrycks said the visual aspect of ARC makes it less suited to assessing language models. "Humanity’s Last Exam" will require abstract reasoning, he said.

Answers from common benchmarks may also have ended up in data used to train AI systems, industry observers have said. Hendrycks said some questions on "Humanity's Last Exam" will remain private to make sure AI systems' answers are not from memorization.

The exam will include at least 1,000 crowd-sourced questions due November 1 that are hard for non-experts to answer. These will undergo peer review, with winning submissions offered co-authorship and up to $5,000 prizes sponsored by Scale AI.

"We desperately need harder tests for expert-level models to measure the rapid progress of AI," said Alexandr Wang, Scale's CEO.

One restriction: the organizers want no questions about weapons, which some say would be too dangerous for AI to study.