Meta faced multiple high-profile AI demo failures at its Connect conference, with the company’s CTO later attributing the incidents to an accidental distributed denial-of-service (DDoS) attack and an obscure software bug. The failures highlighted the challenges facing live AI technology demonstrations and raised questions about the readiness of Meta’s smart glasses technology for widespread deployment.
What happened: Two major demos malfunctioned during Meta’s showcase of its Ray-Ban Meta smart glasses with Live AI capabilities.
- During the first demo, an Instagram influencer attempting to get cooking help from the AI assistant experienced multiple failures, with the AI incorrectly assessing his progress and failing to respond to interruptions.
- CEO Mark Zuckerberg’s demonstration of the neural band wristband technology failed repeatedly when he couldn’t answer incoming WhatsApp calls using finger gestures, despite five attempts.
The technical explanation: Meta CTO Andrew “Boz” Bosworth blamed the failures on an unexpected system overload and a rare bug.
- “When the chef said, ‘Hey Meta, start Live AI’, it started every single Ray-Ban Meta’s Live AI in the building, and there were a lot of people in that building,” Bosworth explained on Instagram Stories.
- The company had routed Live AI traffic to development servers but failed to account for all devices in the venue, essentially launching an accidental DDoS attack against their own systems.
- The call-answering issue resulted from the display going to sleep at the exact moment a notification arrived, preventing the answer tool from appearing even after the display was reactivated.
In plain English: A DDoS (distributed denial-of-service) attack is like hundreds of people trying to enter a store through one door at the same time—it creates a jam that prevents anyone from getting through. Meta accidentally created this situation when every pair of smart glasses in the building activated simultaneously, overwhelming their servers with requests.
What critics are saying: The technical explanations sparked skepticism about Meta’s infrastructure capabilities.
- “How can 10’s of devices DDOS
META’sservers?” one X user questioned. - Another critic called it “the wildest excuse I’ve ever heard” that maybe 100 devices could accidentally overwhelm Meta’s servers.
Bosworth’s response: The CTO remained confident in the underlying technology despite the public failures.
- “Obviously, I don’t love [the demo fails], but I know the product works. I know it has the goods, so it really was just a demo fail and not a product failure,” he said.
- He noted that the call notification bug had never occurred before and has since been fixed.
The broader context: The failures occurred during Meta’s demonstration of ambitious new technologies that face inherent challenges.
- Zuckerberg had warned before the demos that the company is still working through “major technology challenges” with Live AI, particularly regarding all-day availability.
- The Live AI feature is currently limited to one or two hours of use at a time.
- The neural band represents what Zuckerberg called a “huge scientific leap” designed to replace keyboards and mice for smart glasses control.
Recent Stories
DOE fusion roadmap targets 2030s commercial deployment as AI drives $9B investment
The Department of Energy has released a new roadmap targeting commercial-scale fusion power deployment by the mid-2030s, though the plan lacks specific funding commitments and relies on scientific breakthroughs that have eluded researchers for decades. The strategy emphasizes public-private partnerships and positions AI as both a research tool and motivation for developing fusion energy to meet data centers' growing electricity demands. The big picture: The DOE's roadmap aims to "deliver the public infrastructure that supports the fusion private sector scale up in the 2030s," but acknowledges it cannot commit to specific funding levels and remains subject to Congressional appropriations. Why...
Oct 17, 2025Tying it all together: Credo’s purple cables power the $4B AI data center boom
Credo, a Silicon Valley semiconductor company specializing in data center cables and chips, has seen its stock price more than double this year to $143.61, following a 245% surge in 2024. The company's signature purple cables, which cost between $300-$500 each, have become essential infrastructure for AI data centers, positioning Credo to capitalize on the trillion-dollar AI infrastructure expansion as hyperscalers like Amazon, Microsoft, and Elon Musk's xAI rapidly build out massive computing facilities. What you should know: Credo's active electrical cables (AECs) are becoming indispensable for connecting the massive GPU clusters required for AI training and inference. The company...
Oct 17, 2025Vatican launches Latin American AI network for human development
The Vatican hosted a two-day conference bringing together 50 global experts to explore how artificial intelligence can advance peace, social justice, and human development. The event launched the Latin American AI Network for Integral Human Development and established principles for ethical AI governance that prioritize human dignity over technological advancement. What you should know: The Pontifical Academy of Social Sciences, the Vatican's research body for social issues, organized the "Digital Rerum Novarum" conference on October 16-17, combining academic research with practical AI applications. Participants included leading experts from MIT, Microsoft, Columbia University, the UN, and major European institutions. The conference...