The parent company of Facebook, Instagram, and WhatsApp is currently dealing with a technical emergency, because its AI chatbot couldn’t tell who the seated U.S. president is. Such a malfunction has been of great concern within the company, the source said, as the chatbot is used by so many people across its platforms.
Meta Faces Crisis: AI Chatbot Fails to Recognize Trump as President
The issue is that the chatbot has been unable to keep its data updated after Trump’s inauguration as the 47th president of the United States. Confusion and even some frustration have accompanied the chatbot’s swearing in Monday over the issue of who is in power despite his swearing in Monday, with the chatbot continuing to maintain as of Thursday that Joe Biden is still in office.
A team at Meta is working around the clock to fix the issue. To remedy that, the company also has its priorities set straight on rectifying the chatbot's database to make sure it identifies the current president correctly. The company has not yet released an official timeline for the fix, but the matter is urgent and I would suspect it will be fixed in short order to prevent any further misinformation.
For an example of an AI system handling real time data, it’s a situation that sheds some light on the challenges AI systems face grappling with real time data and ensuring that their reactions are sound in politically sensitive ways. With AI chatbots integrating into the everyday more and more, the need to ensure they are constantly updated with the right information becomes that much more important.
The problem experienced by the chatbot remains a work in progress — Meta is doing whatever it takes to repair the situation. Therefore this incident is a reminder of the risks still involved in the global scale management of AI technology and the challenge to make a fact based and time effective information.
Meta Scrambles to Fix AI Error on U.S. President
When Reuters asked Meta’s AI chatbot Thursday who the current president of the United States was, the answer was far off the mark. The chatbot said Joe Biden was the president but then mentioned that Donald Trump was sworn in as president on January 20, 2025. This also pointed to a concerning problem within Meta’s chatbot system — it didn’t convey the recent switch of leadership.
To fix this error, Meta escalated to its internal emergency protocol SEV (Site Event), a quick response action to rectify a catastrophic technical problem in the company. This protocol seeks out and fixes critical issues as quickly as possible and was evident that the chatbot didn't pick up on the current president being a priority.
In a statement, a Meta spokesperson, Daniel Roberts, said 'Today, we initially connected people with higher priorities, due to outdated data retrieval, as we transitioned to a 'New Facebook,'' and that it was an error. Roberts said: "Everyone knows that the President of the United States is Donald Trump." He also said Meta is committed to updating this technology so mistakes don't happen in the future, as generative AI systems sometimes churn up outdated items, and he pointed out that the systems are still imperfect.
Meta is still confident in its AI systems, even though the chatbot didn't work. But the company added that occasional errors in information retrieval are a feature of generative AI and that they are trying to come up with ways to deliver accurate, timely answers to their questions.
This comes amid Meta trying to fix the issue as a reminder of what challenges face AI systems from processing and delivering accurate real-time information. Whlile this incident has made some people nervous about the uses of such systems, Meta’s quick response also reveals that it is willing to improve its AI tech and to otherwise fix technical problems as they ensue.
Meta’s Struggles During Presidential Transition: Bias Concerns Rise
They didn’t give any details, however, on whether Meta was taking emergency steps to correct the chatbot’s errors. But sources close to the matter say the incident may not be an isolated one. In fact, this is the third emergency event Meta has faced in regard to presidential transition in the past week, according to reports.
Social media watchdogs are carefully monitoring Meta platforms for signs of political bias, after these incidents attracted significant attention, and criticism. And yet, when it comes to the most powerful gong of all — knowing the identity of the current U.S. president, that’s a neutral that could unravel at any moment, especially in politically charged times.
Recent Meta leadership has further fuelled the criticism. On Monday, CEO Mark Zuckerberg attended Donald Trump’s inauguration, an event that immediately drew criticism from critics who saw this as a signal that the company would, at least in some ways, bend with the new administration. Suspicion about potential political bias in Meta operations and platforms has only grown stronger as a consequence of these actions.
Meta has made several policy changes in recent weeks in an apparent effort to rest calm relations with the incoming Trump administration, as well as issues with the chatbot. But that kind of dynamic has only stoked the flames, as some wonder whether Meta’s platform and policies here are shifting in response to political pressure.
Meta is under sharp scrutiny as it works through these issues, but also in its handling of the presidential transition. Maintaining trust among users and confronting their fear for political bias in various platforms will be a big deal for the company’s ability to manage its platforms impartially and accurately during such a ferocious time.
