Apple continues to allow fake AI news summaries on iPhones

In a similar vein to my last post, this week's post highlights more (frankly strange) efforts by big tech to control the flow of information with AI tools. Last week, we talked a bit about Meta and their fake BIPOC/Queer AI profiles that they had to roll back for now.

Apple is being caught in a similar vein of issues with their new "AI Summaries" giving incorrect information regarding current events from the news. Some of these summaries are funny, but some of them are genuinely concerning and bring to question how bad actors and/or for-profit companies could (and are!) using AI to spread misinformation.

Today, we'll be talking about some of the concerns that have been brought up, and why these are concerning in what is largely a poor and biased implementation of AI on Apple's part.

Let's dive in.

AI Summaries can't be that bad, can they?

They certainly can be that bad. Here's a handful of examples I found that Apple Intelligence reported with a quick Google Search. (https://www.telegraph.co.uk/news/2025/01/07/apple-told-to-drop-ai-that-pushed-fake-bbc-news-stories/)

• AI claims BBC News reports Luigi Mangione, accused assassin of the former United Health Care CEO, shot himself. Luigi did not shoot himself.

• AI claims New York Times reports Benjamin Netanyahu, prime minister of Israel, had been arrested. If only we were so lucky.

• AI claims BBC News reports famous tennis player Rafael Nadal comes out as gay. He did not.

• AI claims BBC News reports that Luke Litter wins the World Darts Champoinship. The match had not even been played at the time.

If you notice, BBC News is being consistently brought up in these examples. They've noticed this as well. The BBC put a complaint out in December when Apple incorrectly reported the suicide of Luigi Mangione and claimed BBC News was the source.

What is Apple doing about this?

Recently, Apple put out a statement regarding their updates in the "coming weeks" to their AI summary features:

So, there's no clear information on how Apple plans to make AI summaries better. They just are making it easier to tell on their software that a summary was provided by AI. Whatever that means. It's clear that Apple is passing the buck onto the consumer about these summaries in the way they are stating how the function works. Don't like the AI summaries? They're optional. We'll make it clearer for you when they're AI, I guess.

They are also seemingly ignoring calls from organizations like BBC News, ProPublica, the National Union of Journalists (NUJ), and Reporters Without Borders (Reporters sans frontières; RSF) to remove the feature to avoid misinforming the public. Laura Davison, NUJ general secretary, had the following to say:

So where does this leave us?

It frankly leaves us at the whim of a private company to determine what the public is receiving in terms of information. Similarly with concerns about Meta's AI profiles on Facebook and Instagram, Apple's behavior is an absence of ownership to understand the impact releasing a poorly implemented AI feature like this is having to news organizations and the general public.

This lack of ownership on Apple's part creates an incredibly damaging source of misinformation coming directly to iPhone users' inboxes. Users having to opt-out of what is largely an incomplete feature that is falsifying claims under the name of news outlets people trust shows that Apple wants their users to stay misinformed. Otherwise, it would be easy to temporarily disable the feature while they work out the kinks. Or better yet: not release an incomplete product in the first place.

It is also another piece of evidence to suggest that social media companies, and companies that are providing information to the public, should be held to a higher standard. A tough line to walk, since these companies are private and these services are things you can "just disable" or "not use." However, these products like Instagram, Facebook, and Twitter are so permeated into how we communicate on a daily basis that we can't just get off them en masse. An individual certainly could though. The most we've seen is either users flocking from one social media to another (Twitter -> Blue Sky, Twitter -> Threads), or government taking steps to ban a social media platform like TikTok in the U.S. Outright social media abandonment is unfortunately harder than it sounds.

Going back to Apple's AI summaries: I would advise that people look into either disabling the feature, or scrapping your iPhones altogether for another phone operating system that isn't providing inaccurate news summaries. I can't guarantee that a Pixel or Samsung or similar type of phone won't have additional features similar to Apple's fake news summaries, but you might have more luck with an economy Android model from other companies than those three. Reddit can also help:

Conclusion

Be safe out there. There's so many different sources of misinformation coming through as a result of poorly implemented AI solutions, and we should be mindful of the content we're consuming. Much of the onus is on us, sadly, to navigate what is real with features like Apple's AI summary. Sharing and keeping your friends and family informed is a great start.

Thanks for reading. Be back soon to share more same time next week!

Snowfall in Colorado means Cha Cha has to don his booties and clothes to go outside. He's not a fan. The new heated blanket should get a lot of good use.