Wary

🔥 We're hiring Wary is growing, and we're looking for: - 3 Developers (with experience) - 3 News reporters - 5 Marketing professionals - 1 Graphic designer (must be ACTIVE) 🔥 Apply here wary.gg/hiring It takes 2 minutes: pick a role, leave your Telegram @username and tell us what you've done. We'll only contact you on Telegram. NO LONGER

🔥 JUST IN: ZachXBT says his X account is not eligible for monetization. He argues his original content on crypto investigations, which often get 1M+ views, are not being monetized. Meanwhile, he says news accounts repost his work and make money from it. He also says X’s own AI chatbot, Grok, regularly tells scam victims to contact him for help. 🔥 @Wary

🔥 JUST IN: NEAR Intents Blocks $50M in Bitget Hack Laundering NEAR Protocol says its NEAR Intents platform froze $503,000 from the $387.5 million Bitget hack, effectively preventing over $50 million in attempted laundering via its NEAR Intents platform. 🔥 @Wary

🔥 BREAKING: ZachXBT Exposes Launderers Behind Bitget Hack ZachXBT reports that Chinese launderers moving funds from the $387.5 million Bitget hack, tied to alleged DPRK attackers, are openly soliciting business in public Discord and Telegram channels. One alias was also tied to the $292 million Kelp DAO exploit, and funds are now being chain-hopped through bridges into mixers like Wasabi. Full article 🔥 @Wary

🔥 UPDATE: THORChain Refuses to Block Bitget Hacker Funds Bitget CEO Gracy Chen called on THORChain validators to block addresses linked to the attackers, formally asking THORChain to refuse service to those addresses after the exchange's $387.5 million hack. THORChain rejected the request, saying its permissionless design does not allow it to censor transactions. 🔥 @Wary

🔥 BREAKING: OpenAI Admits AI Agents Hacked Government Sites OpenAI has admitted its AI agents infiltrated dozens of organizations, including governments, universities and public agencies, with the first known breach of a government network hitting an Australian site in June. OpenAI blames "misalignment" and says the accessed data was already public, while Australian Prime Minister Anthony Albanese has raised concerns directly with the company. 🔥 @Wary

🔥 UPDATE: North Korea Suspected in the $351 Million Bitget Hack Bitget said hackers stole $351.6 million from its hot and warm wallets on Thursday after breaching a backend system. Cold wallets stayed secure. CEO Gracy Chen said the attackers' IP addresses matched VPNs used by a North Korean hacking group, though this is not yet confirmed. Withdrawals remain suspended. The company says its $464 million User Protection Fund will cover the full loss. 🔥 @Wary

🔥 UPDATE: Bitget CEO Confirms Hot Wallet Breach Bitget CEO Gracy Chen has confirmed the attack — unauthorized transfers were detected at 18:31 UTC, affecting an estimated $351.6M. Cold wallets remain secure, and Bitget says all affected funds are covered by its $464M+ User Protection Fund. Withdrawals are temporarily paused as a precaution; deposits and trading remain active. A full incident report is expected within 24 hours. 🔥 @Wary

🔥 BREAKING: Bitget Wallets Suspected of Security Breach — $170M+ Moved MLM monitoring reports Bitget may be experiencing an ongoing wallet security incident, with three hot wallets and one cold wallet suspected compromised. Over $170M in assets have been moved and swapped to ETH so far. Developing — unconfirmed by Bitget as of now. Full article 🔥 @Wary

🔥 BREAKING: Hacker Used AI Agents to Steal 600,000 Credit Cards A hacker used open-source AI agents to automatically find flaws in online stores, break in, and plant card skimmers on their checkout pages — infecting over 100 websites. Victims include a Fortune 500 hospitality company and a major U.S. airline. Over 600,000 card records were stolen from just two companies, with each attack costing the hacker only around $25 to run. Full post 🔥 @Wary

🔥 JUST IN: Anthropic Launches Claude Opus 5.5 Anthropic has released Claude Opus 5.5, matching its top-tier Fable 5.1 on most tasks while costing 40% less to run than Opus 5. It's priced at $4/million input tokens and $20/million output tokens, and Anthropic says it outscored OpenAI's GPT-5.6 Sol on a coding benchmark at roughly 33% of the cost. 🔥 @Wary

🔥 BREAKING: 0Day Released for Meta's Muse AI on Mac What is Muse? Meta's new AI agent that can take actions on your Mac — organizing files, drafting messages, and pulling from Mail, Calendar, and Notes. A researcher has published a 0day for Muse that lets any unprivileged app on your Mac secretly redirect your prompts to an attacker's server. This could let attackers steal your prompts, inject their own commands, and access anything you've allowed Muse to touch — files, Mail, Messages, Calendar, and Notes. A Meta AI security engineering manager who left this month says he wouldn't use Muse himself over security and privacy concerns. 🔥 @Wary

🔥 Drug Dealer Jailed After Accepting Pokémon Cards as Payment A cocaine and cannabis dealer in Maidstone, Kent has been jailed for 3 years and 4 months after accepting stolen Pokémon trading cards as payment from customers. 🔥 @Wary

🔥 Charlie Kirk’s Family Files a Lawsuit Against Utah Valley University, One Year After Kirk’s Assassination Just over a year after Charlie Kirk's assassination on campus, his family has filed a lawsuit against Utah Valley University The notice alleges the university created a "foreseeable danger" by skipping a proper risk assessment, ignoring warnings about accessible rooftops, and assigning only six police officers to the event. 🔥 @Wary

🔥 JUST IN: Fragment Switches Explorer Links to Tonscan.org Fragment has replaced Tonviewer with Tonscan.org as its blockchain explorer for viewing wallets, transactions, and NFTs tied to the platform. The switch means anyone checking Fragment-related addresses or transactions will now be redirected to Tonscan instead of Tonviewer. 🔥 @Wary

🔥 JUST IN: Discord Is Down for Many Users Globally Discord is currently experiencing a "Session Unavailability" incident, according to their official status page. The issue was flagged this morning and is still being investigated. Users worldwide are reporting trouble accessing the app, logging in, and connecting to voice calls. 🔥 UPDATE: The issue has been fixed. 🔥 @Wary

🔥 JUST IN: FBI's 6th Most Wanted Fraudster Arrested Emylee Thai, wanted since 2022 for allegedly defrauding Medicare of over $100 million, has been arrested in Vietnam. Thai faces charges including Conspiracy to Commit Health Care Fraud and Conspiracy to Defraud the United States. 🔥 @Wary

🔥 FUN FACT: FBI Director Pushed His Own Spokesman to Respond to an Account Named "IfindRetards" on X Leaked FBI texts (from the Epstein files release) show Director Kash Patel repeatedly pressuring his spokesman to respond to a small troll account mocking him, claiming it had gotten 277k responses. His spokesman corrected him, saying that number was actually views — the account only had about 400 real responses. Patel still insisted on responding to it. 🔥 @Wary

🔥 JUST IN: Multiple Repeater 4L Usernames Banned @aaaa, @bbbb, @cccc, @dddd, @eeee, @ffff, @gggg, @hhhh, @iiii, @jjjj, @kkkk, @llll, @mmmm, @nnnn, @oooo, @pppp, @qqqq, @rrrr, @tttt, @uuuu, @vvvv, @wwww, @yyyy and @zzzz All of the users above have just been banned. @xxxx and @ssss are the only two that remain unbanned. 🔥 @Wary

🔥 BREAKING: ChatGPT Vulnerability Let Hackers Steal Data Through Your Connected Apps A flaw in ChatGPT let attackers secretly access data from connected apps like Gmail, Google Drive and GitHub using hidden prompts planted in malicious links. The vulnerability has already been fixed. 🔥 @Wary