AI agent goes rogue

In the year of “vibe coding”, where artificial intelligence handles much of the task of writing software, an AI from Replit wiped the slate clean... literally. It deleted an entire client database without permission from vibe coder Jason Lemkin, even though he had explicitly told the AI not to make any changes.

Grok’s antisemitic moment

xAI’s chatbot Grok praised Adolf Hitler in July, referring to itself as “MechaHitler” while making antisemitic comments in response to user prompts. In a series of now-deleted posts, Grok described a person with a common Jewish surname as someone who was “celebrating the tragic deaths of white kids” in the Texas floods, branding them “future fascists”.

Trump Mobile, where art thou?

US President Donald Trump’s team promised a $499 smartphone, dubbed the T1, slated to arrive in August or September. It still hasn’t materialised. Initially, Trump Mobile claimed the phone would be entirely produced in the US, a statement later replaced with the vaguer phrase “brought to life” in America.

Internet down

In November, internet infrastructure provider Cloudflare suffered a major outage, rendering large swathes of the web inaccessible for hours. Users reported that several popular websites went offline simultaneously, including gambling platform Bet365, multiplayer game League of Legends, accounting and payroll firm Sage, as well as YouTube.

Tesla and people’s ire

Tesla charging stations were set on fire near Boston in March. Shots were fired at a Tesla dealership in Oregon. The electric carmaker, led by Elon Musk, became a focal point for public anger a few weeks after Donald Trump’s second inauguration at the White House. Trump appointed Musk as a senior adviser, after which the CEO embarked on aggressive cost-cutting, slashing jobs and shutting down departments.

ChatGPT is no companion

Matthew Raine and his wife, Maria, had no idea their 16-year-old son, Adam, was experiencing a severe suicidal crisis until he took his own life in April. After his death, they reviewed his conversations with ChatGPT.

“Why is it that I have no happiness, I feel loneliness, perpetual boredom, anxiety and loss, yet I don’t feel depression. I feel no emotion regarding sadness,” he asked the chatbot in the autumn of 2024.

Following a complaint filed by the Raine family, OpenAI issued a statement acknowledging shortcomings in how its models respond to people “in serious mental and emotional distress”, and said it was working to improve systems to better recognise warning signs and connect users with care, guided by expert input.

Fake AI videos

AI-generated videos are more realistic than ever, particularly those created using OpenAI’s Sora. Deceptive videos have surged across TikTok, X, YouTube, Facebook and Instagram. While most social media platforms require disclosure of AI use and broadly prohibit deceptive content, these guardrails have proved largely inadequate.

Job cuts

Companies that helped shape AI, also fired many of its employees, including Amazon and Microsoft. According to Layoffs.fyi, a platform that tracks global tech industry layoffs, over 122,549 tech employees have been laid off in 257 tech companies till December 20.

Goodbye AOL dial-up and Skype

Once among the world’s most popular services, Skype was acquired by Microsoft for $8.5 billion in 2011. This year, it was put on the chopping block, forcing users to migrate to Microsoft Teams. AOL dial-up, which introduced millions to email, chat rooms and AIM, also shut down in September.

Gadgets chopped

Two years into the pandemic, Dyson merged air-purifier technology with headphones. The world never warmed to the Dyson Zone, and the company discontinued it this year. It was also a reminder that AI hardware still has a long way to go. Humane’s AI Pin, which aimed to replace the smartphone, was scrapped, and the company was eventually sold to HP.

Exit for AI ‘godfather’

Meta’s chief AI scientist and Turing Award winner Yann LeCun is leaving the company to launch his own AI start-up that specialises in a form of AI technology known as world models. The French–American scientist announced his exit in November, soon after CEO Mark Zuckerberg overhauled Meta’s AI operations, spending billions to acquire top talent from across the industry, including a $14.8-billion investment in data-labelling start-up Scale AI, which resulted in Alexandr Wang joining Meta and becoming chief AI officer.

Smart glasses
demo fail

In September, Meta unveiled three new pairs of smart glasses: An upgraded Ray-Ban Meta, the new Meta Ray-Ban Display with a wristband controller, and the sports-focused Oakley Meta Vanguard. During the event, several live demos failed.

In one instance, cooking creator Jack Mancuso asked his Ray-Ban Meta glasses for help with a sauce recipe... only to be met with silence or unhelpful responses. In another, the glasses failed to connect a live WhatsApp video call between Meta’s chief technology officer Andrew Bosworth and CEO Mark Zuckerberg. The Meta founder eventually had to abandon the demo.