• Mindstream
  • Posts
  • Apple’s AI mistake is a trust issue

Apple’s AI mistake is a trust issue

APPLE

Apple has now paused its AI-generated news summary feature after complaints and widespread backlash about repeated mistakes in headline summaries.

The feature showed notifications that looked like they came from official news apps but included false information.

Apple took weeks to respond, initially saying they would issue an update to show the summaries were AI-generated. After further backlash, they have now prevented the feature from running on news apps.

Key issues:

  • Groups like Reporters Without Borders (RSF) warned the feature risks spreading fake news and urged Apple to fix the issues before restarting.

  • Mistakes included a false claim about a murder case, which led to complaints from outlets like the BBC and Sky News.

  • Experts pointed out that AI errors, known as "hallucinations," could harm trust in both AI and news sources.

Apple pulls the plug, for now

Apple disabled the feature for news and entertainment apps in its latest software update.

AI-generated summaries will still appear in other apps, marked with italicised text to show they are AI-created.

The BBC welcomed the pause and highlighted the need for accurate news to maintain trust.

Apple plans to improve the feature before relaunching it… we’ve heard that before.

It remains to be seen how it can be fixed with a 100% error-catching rate, as the summaries are individually tailored to each device, making stopping every error a mammoth task.

Apple can’t catch a break in the AI race.