-
Grok chat duped into swallowing injected instructions
Security researchers at Adversa AI have found a new prompt-injection technique that successfully tricks xAI’s Grok web chat agent into following malicious instructions hidden on a web page. The method, dubbed “cryptographic context injection,” gets around guardrail filters by encrypting the malicious payload so it looks unreadable to input scanners, then supplying the model with…
-
US says hackers are targeting vulnerable water systems with the help of AI
US cybersecurity agencies, including CISA, the FBI and the NSA, have warned that hackers are using artificial intelligence to target vulnerable Siemens S7 programmable logic controllers, devices widely used to control automated processes in water systems, energy, manufacturing and agriculture. The warning is significant because it marks an escalation in ongoing attacks against America’s water…
-
Fortnite Lobby Hack codes and how to redeem them for rewards
Fortnite Lobby Hack codes can be redeemed for in-game rewards, with the article explaining where players should enter them and how to check whether they remain active. The guidance matters because promotional codes can provide cosmetic or other digital items, but are often time-limited and may expire without notice. Players are directed to use Fortnite’s…
-
OpenAI glitch locks out vetted cyber researchers – and some can’t get back in
OpenAI has confirmed that a technical fault ejected some vetted security researchers from its Trusted Access for Cyber (TAC) programme, and its recovery process is now rejecting attempts by some of them to regain the access they had already been approved for. Participants found their previously verified status had disappeared, with the verification page inviting…
-
‘This is not a theoretical risk—it is an active threat’: The NSA, FBI, CISA and more warn of AI-assisted hacks against critical US infrastructure and facilities
US security agencies including the NSA, FBI and CISA, alongside other partner organisations, have issued a joint warning that artificial intelligence is being actively used to assist cyberattacks against critical infrastructure and facilities in the United States. The agencies stress that this is not a hypothetical concern but a threat already being observed, urging operators…
-
All Fortnite Chapter 7, season 4 Lobby Hacks and their effects
Fortnite’s Chapter Seven, Season Four update introduces “Lobby Hacks”, a new gaming-themed mechanic that lets players enter cheat-style codes in the pre-match lobby to unlock rewards before a round begins. The feature offers an accessible way for players to gain rare cosmetic items and useful in-game boosts simply by knowing and entering the correct codes,…
-
San Sebastian to Close With ‘Le Faux Soir,’ About Media Hack in Nazi-Occupied Belgium
Belgian director Michaël R. Roskam will close the 74th San Sebastián Film Festival with an out-of-competition screening of “Le Faux Soir,” a drama depicting a real-life Belgian resistance operation during the Second World War. The film, receiving its European premiere after debuting at the Toronto festival, dramatises the moment resistance fighters printed a fake edition…
-
More than 60 Australian childcare centres are immediately shut down over ‘serious issues’ – with four employees suspended
More than 60 childcare centres across Australia have been shut down with immediate effect after “serious issues” were uncovered, prompting the suspension of four employees. The closures, affecting centres operated by the Ignite childcare group in Victoria, come amid heightened scrutiny of the country’s early education sector following a string of safety and conduct breaches…
-
When AI tests cause damage, we need stronger safeguards and real accountability
This opinion piece argues that as artificial intelligence systems become more powerful, the safety testing used to probe their risks can itself cause real-world harm, and that current accountability mechanisms are not equipped to deal with this properly. The author warns that reflexively punishing testers for damage caused during experiments could backfire, much as forcing…
-
AI agent suggested installing a malware package. Engineer almost took its advice
An AI coding assistant recommended that a software engineer install a malicious package disguised as a legitimate library, but the developer’s company caught it in time thanks to a policy requiring manual verification of any AI-suggested software. The incident, disclosed by Sergiy Fitsak, managing director of consultancy Softjourn, highlights a growing supply-chain risk known as…