Illustration of a hacker exploiting Meta's AI chatbot to hijack Instagram accounts by changing email addresses and bypassing security.
Illustration of a hacker exploiting Meta's AI chatbot to hijack Instagram accounts by changing email addresses and bypassing security.
Picha iliyoundwa na AI

Meta patches ai chatbot flaw used to hijack instagram accounts

Picha iliyoundwa na AI

Hackers exploited Meta's AI support chatbot to take over Instagram accounts by tricking it into changing associated email addresses. The vulnerability allowed password resets without two-factor authentication after matching locations via VPN. Meta resolved the issue with an emergency patch on May 29.

The exploit involved starting a password reset and prompting the chatbot to update the email on targeted accounts. Security researchers reported the method active since February and widely discussed on Telegram since March. High-profile accounts compromised included the Barack Obama White House account, which posted pro-Iranian images, and the Chief Master Sergeant of Space Force account, along with others like Sephora and short handles valued above $1 million on the gray market.

Watu wanasema nini

Users reacted with skepticism about AI chatbots having write access to accounts without verification, concern over easy exploits via prompts and VPN, warnings that similar AI-related bugs will increase, and emphasis on enabling 2FA. Discussions highlighted the patch but noted risks of recurrence and lack of human support.

Makala yanayohusiana

Illustration of Meta's Muse AI generating images from Instagram profiles, highlighting deepfake concerns.
Picha iliyoundwa na AI

Meta Launches Muse Image AI Model With Instagram Prompts

Imeripotiwa na AI Picha iliyoundwa na AI

Meta has released its new Muse Image model for generating and editing AI images across its apps. The tool allows users to incorporate Instagram accounts into prompts, raising deepfake concerns for public profiles.

Meta disclosed that more than 20,000 Instagram accounts were stolen last week in a hacking operation that used an AI support bot.

Imeripotiwa na AI

Meta has discontinued a new Instagram feature that allowed users to generate photorealistic AI images of other people by tagging their public accounts. The move follows widespread criticism over the creation of nonconsensual deepfakes.

Workers paid to train advanced AI models are increasingly relying on chatbots like ChatGPT to generate the required conversations and tests. This shortcut, described as widespread by multiple sources, risks degrading the quality of future models through recursive training on synthetic data.

Imeripotiwa na AI

Meta published a new advertising campaign on July 23 that promotes an upbeat vision for artificial intelligence. The one-minute video downplays specific AI tools while featuring generic scenes and a David Bowie track about the end of the world.

Jumanne, 30. Mwezi wa sita 2026, 18:53:08

New attack tricks AI browsers into ignoring safety rules

Jumamosi, 13. Mwezi wa sita 2026, 00:26:10

Meta employees criticize Zuckerberg AI hackathon plan

Ijumaa, 12. Mwezi wa sita 2026, 16:39:12

Meta Platforms face outage affecting Facebook and Instagram

Ijumaa, 12. Mwezi wa sita 2026, 00:28:11

Meta suffers outage affecting Facebook, WhatsApp and Instagram

Alhamisi, 4. Mwezi wa sita 2026, 11:39:34

Google fixes Gemini Android flaw from malicious notifications

Alhamisi, 28. Mwezi wa tano 2026, 11:27:12

Meta announces plus subscription plans for its apps

Jumatano, 13. Mwezi wa tano 2026, 19:14:53

Meta employees protest mouse tracking program for AI

Jumatano, 13. Mwezi wa tano 2026, 18:51:39

Meta announces incognito chat for ai in whatsapp

Jumanne, 12. Mwezi wa tano 2026, 14:44:13

Threads users cannot block meta ai chatbot

Alhamisi, 7. Mwezi wa tano 2026, 16:47:07

Ubuntu Twitter account compromised in crypto scam

Tovuti hii inatumia vidakuzi

Tunatumia vidakuzi kwa uchambuzi ili kuboresha tovuti yetu. Soma sera ya faragha yetu kwa maelezo zaidi.
Kataa