Connect with us

Published

on

Apple CEO Tim Cook delivers remarks at the start of the Apple Worldwide Developers Conference (WWDC) on June 10, 2024 in Cupertino, California. Apple will announce plans to incorporate artificial intelligence (AI) into Apple software and hardware. (

Justin Sullivan | Getty Images

Apple has temporarily disabled Apple Intelligence summaries for news apps for users of its beta software in a sign of the challenges the company is dealing with in its AI technology. 

The decision to pause AI summaries comes weeks after the BBC highlighted that Apple’s AI system had twisted its news notifications to display inaccurate facts. The pause only affects people using Apple’s beta software, not those using the company’s main operating systems. 

News and entertainment apps, such as The New York Times, began showing a short message inside the iPhone settings app on Thursday noting that AI-powered summaries were “temporarily unavailable.”

The pause on one of Apple Intelligence’s core features highlights the challenge Apple faces in the roll out of its artificial intelligence technology, which has been scrutinized by many users on social media.

“With the latest beta software releases of iOS 18.3, iPadOS 18.3, and macOS Sequoia 15.3, Notification summaries for the News & Entertainment category will be temporarily unavailable,” an Apple representative told CNBC in a statement. 

The spokesperson noted that Apple is working on improvements to the software that are coming in a future software update. The company did not say when it will roll out its iOS 18.3 software to users of the main version of the iPhone operating system, but it could take weeks, based on Apple’s previous software release patterns.

The decision to temporarily pause the AI summaries comes on the same day that Apple saw its stock close down 4%, marking its worst day of trading since Aug. 5. A reason for the drop was due to notable Apple supply chain analyst Ming-Chi Kuo writing on Monday that the Apple Intelligence suite of features does not appear to be boosting iPhone sales.

Apple Intelligence struggles since launch

The company launched Apple Intelligence in October as the signature feature in its latest line of iPhone models and its answer to Silicon Valley’s AI arms race that kicked off with the launch of OpenAI’s ChatGPT in late 2022. 

Apple has used the AI features as the key selling point in its advertisements and marketing for its latest hardware products, but the software has been riddled with issues. 

The company says that the entire Apple Intelligence system is in beta, and the update on Thursday added language to say that the AI software can produce unexpected results. 

Apple Intelligence includes several features, including image generators, but the one that’s received the most attention is how it can summarize entire stacks of notifications into concise sentences — useful, according to Apple’s marketing materials, for getting through hundreds of group chat notifications without scrolling through the whole discussion.

With the Thursday update, Apple said it will show any AI-summarized notification in italics to distinguish them from other notifications. 

In testing, Apple Intelligence summaries weren’t perfect, but the errors were mostly funny and obvious. Problems cropped up when the technology began being used to summarize news, and it displayed false information.

The most egregious well-documented error happened in December, when 22 separate BBC news notifications were combined into a three-part headline that started with “Luigi Mangione shoots himself.” The alleged Brian Thompson assassin has not done that.

The feature also combined headlines from The New York Times into a November notification that falsely said that Israeli Prime minister Benjamin Netanyahu had been arrested, according to a ProPublica reporter’s post on social media.

Another Apple Intelligence notification on Jan. 3 said that darts player Luke Littler had won a world championship, which had yet to take place, according to the BBC. The technology also conflated notifications from BBC’s sports app to say that “Brazilian tennis player, Rafael Nadal, comes out as gay.” Nadal is Spanish and is married to Maria Francisca Perello.

Apple on Thursday also rolled out a new feature that lets users turn off AI summaries for any app by swiping left on the notification from the phone’s lock screen. Users previously could only turn off AI summaries through the settings app. 

WATCH: Tariffs, AI and China haunt Apple

Tariffs, AI and China haunt Apple

Continue Reading

Technology

Top EU official denies softer approach to Big Tech, cites ‘very clear legal basis’ for regulation

Published

on

By

Top EU official denies softer approach to Big Tech, cites 'very clear legal basis' for regulation

European Commission is 'fully enforcing' its tech regulations, executive vice president says

A leading EU official has denied taking a softer approach to Big Tech, citing a “very clear legal basis” for regulators and pointing to several ongoing investigations into the likes of social media platform X and Meta.

The FT reported earlier this week that the EU was reassessing investigations into Apple, Google and Meta — a process that could ultimately lead to the European Commission, the executive arm of the EU, scaling back or changing the focus of their probes.

However, speaking to CNBC on Thursday, Henna Virkkunen, the European Commission’s executive vice president for tech sovereignty, pushed back.

“We have our Digital Service Act that came into force a little bit more than one year ago, and there is several formal proceedings going on against, we can say, all the big platforms: Meta platforms, Instagram, Facebook, also on X and with TikTok,” Virkkunen said.

“We are continuing the work, so there is not any new decisions made. So we are doing the investigations [to see] if they are complying with our rules,” she said.

The Digital Services Act or DSA, which came into full effect in 2024, gives EU institutions the power to regulate Big Tech in a bid to prevent illegal and harmful activities online, and clamp down on disinformation.

Despite these new powers, however, there are growing questions about how the EU is actually going to enforce the rules, particularly in the aftermath of President-elect Donald Trump’s return to the White House.

“It remains to be seen what the EU will do, as some investigations have gone further than others, but it is also clear that U.S. tech companies will try to use the Trump administration to push back on EU rules,” Dexter Thillien, lead analyst at the Economist Intelligence Unit, told CNBC.

It comes as the tech industry attempts to cozy up to Trump ahead of his second term as president. Tesla’s Elon Musk, Amazon’s Jeff Bezos and Zuckerberg will attend Trump’s inauguration next week, according to NBC news.

Meta’s CEO Mark Zuckerberg last week, meanwhile, called on the incoming U.S. president to look at the EU’s approach to Big Tech, saying the way the bloc applies competition rules is “almost like a tariff.”

EU official Virkkunen is one of a new team of politicians that began their work as members of the EU’s executive arm in December. Until now, the bloc has been considered a leader of tech regulation and has opened the door to several probes into the behavior of Big Tech companies.

When asked if she was considering taking a softer approach to the sector, Virkkunen said: “We [have a] very clear legal basis and regulation rules in Europe, and of course, now we are fully enforcing those rules.”

Virkkunen did not say whether she was feeling pressure as a result of Trump’s return to the White House. Instead, she said, “all companies, whether American, European or Chinese, have to respect the EU’s regulations.”

Investigating X

In December 2023, Musk’s X was hit with the EU’s first probe under the Digital Services Act. The European Commission is assessing whether X breached transparency obligations and its duties to counter illegal content.

At the time, the institution said it was specifically assessing areas linked to risk management, content moderation, dark patterns, advertising transparency and data access for researchers.

As Musk continues to court the far-right ahead of an election in Germany — including hosting a live discussion with AfD party leader Alice Weidel — there are questions about whether the European Commission will assess this conversation as part of the investigation.

“This is not about Elon Musk. It’s about X,” Virkkunen said.

“X is [a] very large online platform, they have to take their responsibilities, and they have to assess and mitigate the risks, for example, what they are posting for the electoral processes and for civic discourse. But [the European] commission is already investigating X on this, and the scope of investigation is already quite large,” she said, adding that “we are all the time monitoring” in case of new developments.

Continue Reading

Technology

Duolingo shares climb 7% as users swarm to app to learn Mandarin

Published

on

By

Duolingo shares climb 7% as users swarm to app to learn Mandarin

A Duolingo logo is seen on a smartphone.

Pavlo Gonchar | LightRocket | Getty Images

Duolingo shares rose nearly 7% on Thursday following a large spike in users signing up to learn Mandarin in conjunction with soaring usage of Chinese social media app RedNote, a TikTok rival.

The company confirmed to CNBC that there’s been a 216% increase in Mandarin learners using the app compared to a year earlier. For context, Spanish, one of the most popular languages on the app, has seen a 40% increase over the same period, Duolingo said.

RedNote, or Xiaohongshu, as it’s known in China, has rocketed to become the No. 1 free app on the Apple app store, a position it’s held for most of this week. Rounding out the top five are TikTok’s Lemon8 app, U.S. social media upstart Clapper, OpenAI’s ChatGPT and Meta’s Threads.

Last week, the Supreme Court heard oral arguments in the case involving the future of TikTok in the U.S., and a law that could effectively ban the popular app. The justices appeared to favor upholding the law, and a decision could come as soon as Friday. TikTok is reportedly preparing for a U.S. shutdown on Sunday.

RedNote has so far been the top beneficiary of the American user exodus, seeing its U.S. app downloads increase by 20 times over the last week, according to market intelligence firm Sensor Tower. A Duolingo spokesperson told CNBC that the company’s marketing team is “forward-thinking and already has an active presence on Red, managed by our team in China.”

Duolingo offers online and mobile courses across 42 languages. According to its website, Duolingo has 48.8 million Spanish learners. French is the second most popular language on the app at 27.3 million users, while Chinese is eighth at 10.7 million.

Duolingo shares climbed 43% last year, topping the Nasdaq’s 29% gain.

WATCH: AI will continue to be catalyst for Duolingo

AI continues to be catalyst for Duolingo but 'killer' for Shutterstock: Citizens JMP's Andrew Boone

Continue Reading

Technology

Snap shares drop as FTC refers MyAI chatbot complaint to the DOJ

Published

on

By

Snap shares drop as FTC refers MyAI chatbot complaint to the DOJ

Snap CEO Evan Spiegel, joins CNBC ‘Power Lunch’ on September 17, 2024.

CNBC

Snap shares closed down 5% on Thursday after the Federal Trade Commission said it would refer a complaint against the company to the Department of Justice.

The FTC’s non-public complaint involves allegations that Snapchat’s My AI chatbot poses “risks and harms to young users,” the commission said in a statement. The complaint stems from the FTC’s compliance reviews with Snap following a 2014 settlement regarding allegations of public deception pertaining to data collection by the company.

As part of the FTC’s compliance reviews of Snap, the agency said it had uncovered the possibility that the company “is violating or is about to violate the law.”

“A proceeding is in the public interest,” the FTC said in its statement.

The FTC did not specify what about the My AI chatbot its complaint was focused on, but the chatbot has previously drawn scrutiny.

A Snap spokesperson pushed back against the FTC’s claims in a statement to CNBC.

“Unfortunately, on the last day of this Administration, a divided FTC decided to vote out a proposed complaint that does not consider any of these efforts, is based on inaccuracies, and lacks concrete evidence,” the Snap spokesperson said. “It also fails to identify any tangible harm and is subject to serious First Amendment concerns.”

The spokesperson added that while the company shares the FTC’s “focus on ensuring the thoughtful development of generative AI,” Snap believes that the “complaint would stifle innovation and competition in a critical and growing sector of the economy.”

Jonathan Raa | AP

Snap debuted the My AI chatbot in 2023. It is powered by the large language models of OpenAI and Google, giving it the ability to answer user questions and provide tips and suggestions similar to ChatGPT and other AI-powered chatting tools.

The chatbot has been noted for providing problematic responses. In one instance while speaking with a reporter who was pretending to be a teenager, the chatbot answered explained how to hide the smell of alcohol and marijuana, The Washington Post reported in 2023. At the time of the chatbot’s initial release, Snap said that My AI, like other AI-powered chatbots, is “prone to hallucination and can be tricked into saying just about anything. Please be aware of its many deficiencies and sorry in advance!”

In Oct. 2023, the United Kingdom’s Information Commissioner’s Office issued a preliminary enforcement notice against Snap, alleging that the company’s My AI-related risk assessment “did not adequately assess the data protection risks posed by the generative AI technology, particularly to children.”

Although the FTC said that it voted during a closed meeting to issue a public statement about it’s case against Snap and its ensuing referral to the DOJ, it noted that FTC commissioners Melissa Holyoak and Andrew Ferguson were absent.

The FTC also pointed to a dissenting statement by Ferguson, who President-elect Donald Trump named in December to replace Lina Khan as the next FTC chair.

Ferguson noted that these kinds of referrals “are not disclosed unless and until the complaint is filed in court by the Department or the Commission.”

“I did not participate in the farcical closed meeting at which this matter was approved,” he wrote.

Ferguson added that he opposes the FTC’s complaint against Snap, but that he can’t “release a detailed analysis of its many problems,” because the case is not public. Ferguson wrote that the complaint’s interpretations of an FTC law is “wrong” and that it is “in direct conflict with the guarantees of the First Amendment.”

If the DOJ files the complaint, Ferguson said he will “release a more detailed statement about this affront to the Constitution and the rule of law.”

Watch: Snap CEO on earnings beat and new advertising products

Snap CEO on earnings beat and new advertising products

Continue Reading

Trending