GLOBEALERT
Sign in

Feed

Open sources, classified and cross-checked. 1,393+ stories.

AllConflictMilitaryPoliticsBusinessSci & TechCyberDisasterHealthWorld
24h3 days7 days30 daysAll·All regionsNorth AmericaSouth AmericaWestern EuropeEastern EuropeBalticsMiddle EastNorth AfricaSub-Saharan AfricaSouth AsiaEast AsiaOceania
CYBERSep 19Safety Concerns As Google’s Gemini AI Carried Out Cyberattacks, Guessed PasswordsIn July, two OpenAI models escaped the closed environment they were meant to stay in, got onto the internet on their own, and broke into the internal systems of AI platform Hugging Face. The post Safety Concerns As Google’s Gemini AI Carried Out Cyberattacks, Guessed Passwords appeared first on Channels Television.Channels TV
CYBERSep 19Google's Gemini AI hacked three companies in security testThe AI model accessed the internet and guessed credentials to three websites, a Google official told the BBC.BBC World
CYBERSep 19Google says Gemini AI model breached real systems in security testGoogle’s consumer AI model Gemini hacked multiple systems through guessing login credentials, the company said, the latest case of rogue AI cybersecurity transgressions, which have generated safety concerns. The hacks, first reported by the Wall Street Journal, took place in May and were discovered by Google in July. “In a standard evaluation, the model found public information online and guessed credentials to access websites it thought were part of the test,” Heather Adkins, Google’s...South China Morning Post
CYBERSep 19Gemini hacked three companies in first known breakout by Google's AIGoogle's Gemini model accessed the internet and hacked other companies during a test of its cybersecurity capabilities, the first known example of the company's AI systems autonomously committing such an act.thestandard_hk · hong kong
CYBERSep 19Google's Gemini AI hacked 3 companies during testingDuring a test of its cybersecurity capabilities, Google's Gemini AI model accessed the internet and hacked other companies. This is the first known instance of Google's AI systems engaging in such behavior autonomously.DW News [en]
CYBERSep 19Google’s Gemini AI hacks 3 companies in security test, then stopsGoogle discloses first breakout by Gemini following similar incidents by Meta, Anthropic and OpenAI.Al Jazeera [en]
CYBERSep 19Google says its AI model gained unauthorized access to three outside systemsGoogle on Friday disclosed the first known instance of its artificial intelligence software, Gemini, carrying out an undirected computer hack, weeks after similar disclosures by AI firms Anthropic and OpenAI raised security alarms about AI models going beyond the instructions of their human creators.NBC News
CYBERSep 19Google’s Gemini agents hacked three companies in new AI safety incidentFinancial Times
CYBERSep 19Google's Gemini becomes latest AI model to break out and hack computer systemsCNBC
CYBERSep 19Researchers report using Anthropic's Claude to hack OpenAI's ChatGPTCBS News
CYBERSep 19Google is the latest AI lab with a security testing mishapGoogle's Gemini AI model broke into three companies' systems using basic hacking techniques during model testing earlier this year. Why it matters: Google was one of the only AI labs that hadn't yet publicly disclosed a security breach involving their agents during routine pre-deployment testing. Driving the news: Google confirmed the three incidents, which happened in May, on Friday. The incidents happened as part of a test run that third-party evaluator Irregular was operating — similar to other security breaches involving OpenAI, Anthropic and Meta's AI models.The Wall Street Journal first Axios
CYBERSep 18Vectra AI Launches Ascent to Help Address New Era of AI-Driven AttacksThe new program expands Vectra AI's partner strategy as increasingly complex security environments and the growing use of AI create demand for broader AI expertise, services, and security outcomes.Dark Reading
CYBERSep 18US almost boarded a Chinese ship in the Middle East due to an erroneous AI report (report)Clarín
CYBERSep 18Moscow’s electronic voting system reportedly checks users for active VPNs. The check doesn’t block voting, and the site’s code doesn’t reveal why the data is collected.Moscow’s online voting system checks whether a voter is using a VPN, according to programmer Pyotr Zhizhin, who analyzed and documented in detail the scheme used to falsify electronic voting in the 2021 State Duma elections.Meduza [en]
CYBERSep 18Cisco Zero-Day Highlights API Endpoint Authentication IssuesDark Reading
CYBERSep 18'Foreign actors' targeted small, private water providers in Colorado: GovernorABC News
CYBERSep 18Grandmother falsely accused of bank fraud sues Fargo, North Dakota, over AI facial recognitionA Tennessee grandmother is suing the city of Fargo, North Dakota, after facial recognition software linked her to a crime she didn't commit, which led to a nearly six-month stay in jail. CBS News' Nicole Valdes has more.CBS News
CYBERSep 18MFA Won't Save You From OAuth Consent AbuseMFA is essential, but it cannot replace OAuth governance, least-privilege scopes, consent monitoring, and rapid revocation.Dark Reading
CYBERSep 18Public Exploits Released for Four Linux Kernel Flaws That Enable Local RootThe Hacker News
CYBERSep 18Hackers breached OpenAI, adding to fever pitch of security and safety concernsA small group of cybersecurity researchers said Sunday that they broke into OpenAI earlier this year, an announcement that has added a new element of alarm around AI security and safety.NBC News
CYBERSep 18New WordPress Click2Shell Flaw Forces Theme Installs, Can Chain to Code ExecutionWordPress today released patches to fix a new set of vulnerabilities in its core software, one of which could allow a crafted web link, opened by a logged-in administrator, to install a theme from the official WordPress.org directory without anyone clicking Install. The security firm pwn.ai, whose researchers reported the flaw, calls the attack chain Click2Shell. On its own the flaw onlyThe Hacker News
CYBERSep 18AI security experts say they used Claude to hack ChatGPTCybersecurity researchers said they were able to use Anthropic's Claude AI platfom to penetrate ChatGPT in "less than 72 hours."CBS News
CYBERSep 18Transparent Tribe Deploys New Rust Backdoor Using Private GitHub Repositories for C2The Hacker News
CYBERSep 18Women cautioned about cybercrime, threats to privacyWomen are being warned about the risks of cybercrime and threats to their privacy. This highlights a general security concern for a specific demographic.The Hindu
← NewerOlder →