Tag: chatgpt
-
LegalPwn Attack Tricks AI Tools Like ChatGPT and Gemini into Running Malicious Code
Security researchers have discovered a new type of cyberattack that exploits how AI tools process legal text, successfully tricking popular language models into executing dangerous code. Cybersecurity firm Pangea has unveiled a sophisticated attack method called >>LegalPwn
-
LegalPwn Attack Tricks GenAI Tools Into Misclassifying Malware as Safe Code
A new security flaw, LegalPwn, exploits a weakness in generative AI tools like GitHub Copilot and ChatGPT, where malicious code is disguised as legal disclaimers. Learn why human oversight is now more critical than ever for AI security. First seen on hackread.com Jump to article: hackread.com/legalpwn-attack-genai-tools-misclassify-malware-safe-code/
-
Anthropic says OpenAI engineers using Claude Code ahead of GPT-5 launch
Anthropic says it has revoked OpenAI’s access to the Claude API after ChatGPT’s engineers were found using Claude’s coding tools. First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/artificial-intelligence/anthropic-says-openai-engineers-using-claude-code-ahead-of-gpt-5-launch/
-
OpenAI removes ChatGPT self-doxing option
Checkbox to make chatbot conversations appear in search engines deemed a footgun First seen on theregister.com Jump to article: www.theregister.com/2025/08/01/openai_removes_chatgpt_selfdoxing_option/
-
OpenAI may be testing a cheaper paid plan for ChatGPT
OpenAI is reportedly working on a new plan called ‘Go,’ which would be cheaper than the existing $20 Plus subscription. First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/artificial-intelligence/openai-may-be-testing-a-cheaper-paid-plan-for-chatgpt/
-
Search Engines Are Indexing ChatGPT Chats, Here’s What Our OSINT Found
A significant privacy breach has emerged in the artificial intelligence landscape, as ChatGPT shared conversations are being indexed by major search engines, effectively transforming private exchanges into publicly discoverable content accessible to millions of users worldwide. This discovery has exposed thousands of supposedly confidential conversations, ranging from personal mental health discussions to sensitive business information.…
-
Spielerisch gehackt: Forscher umgeht ChatGPT-Schutz und kassiert echte Windows-Lizenzen
First seen on t3n.de Jump to article: t3n.de/news/forscher-umgeht-chatgpt-schutz-und-kassiert-echte-windows-lizenzen-1696521/
-
Browser Extensions Can Exploit ChatGPT, Gemini in ‘Man in the Prompt’ Attack
Man in the Prompt attack shows how browser extensions can exploit ChatGPT, Gemini and other AI tools to steal data or inject hidden prompts. First seen on hackread.com Jump to article: hackread.com/browser-extensions-exploit-chatgpt-gemini-man-in-the-prompt/
-
ChatGPT und Co. via DuckDuckGo – Duck.ai bewahrt Anonymität bei Chatbot-Anfragen
Tags: chatgptFirst seen on security-insider.de Jump to article: www.security-insider.de/duckai-anonyme-nutzung-ki-bots-duckduckgo-a-300ef7a14811943241bca2ab7ed69cf3/
-
ChatGPT, GenAI Tools Open to ‘Man in the Prompt’ Browser Attack
A brand-new cyberattack vector allows threat actors to use a poisoned browser extension to inject malicious prompts into all of the top generative AI tools on the market, including ChatGPT, Gemini, and others. First seen on darkreading.com Jump to article: www.darkreading.com/vulnerabilities-threats/attackers-use-browser-extensions-inject-ai-prompts
-
ChatGPT Agent Defeats Cloudflare’s ‘I Am Not a Robot’ Security Check
In a significant development that highlights both the advancing capabilities of AI and potential vulnerabilities in web security systems, a ChatGPT-powered agent has successfully bypassed Cloudflare’s widely-used >>I am not a robot
-
OpenAI prepares GPT-5 for roll out
OpenAI’s ChatGPT-5 could drop in the coming days, and it could be one of the best models from the Microsoft-backed startup. First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/artificial-intelligence/openai-prepares-gpt-5-for-roll-out/
-
OpenAI could rival Google Shopping with ChatGPT Shop
AI companies like OpenAI and Perplexity like to be the “everything company,” and OpenAI’s latest ChatGPT feature, “Shopping,” makes that obvious. First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/artificial-intelligence/openai-could-rival-google-shopping-with-chatgpt-shop/
-
OpenAI confirms ChatGPT Agent is now rolling out for $20 Plus users
ChatGPT Agent is now rolling out to users with $20 Plus subscription, but OpenAI warns that it will take a few days for the rollout to finish. First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/artificial-intelligence/openai-confirms-chatgpt-agent-is-now-rolling-out-for-20-plus-users/
-
Can ChatGPT Agent Actually Deliver on Its Promises?
OpenAI’s New Agent Automates Tasks, Amid Limits and Privacy Concerns. OpenAI’s new ChatGPT Agent can code, browse and send email. The agent excels at tightly-scoped, well-structured workflows like finding names, drafting content or automating click-heavy tasks, but struggles with ambiguity, creativity or judgment-heavy assignments. First seen on govinfosecurity.com Jump to article: www.govinfosecurity.com/chatgpt-agent-actually-deliver-on-its-promises-a-29045
-
ChatGPT is rolling out ‘personality’ toggles to become your assistant
OpenAI is rolling out a new “personality” feature on the ChatGPT web app. This allows you to choose between multiple personalities, such as “Robot.” First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/artificial-intelligence/chatgpt-is-rolling-out-personality-toggles-to-become-your-assistant/
-
OpenAI confirms ChatGPT’s new study feature, helps with exams
OpenAI is testing a new ‘Study together’ feature, and today, a new announcement within the ChatGPT web app confirms it. First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/artificial-intelligence/openai-confirms-chatgpts-new-study-feature-helps-with-exams/
-
Forscher täuscht ChatGPT mit Ratespiel und ergattert echte Windows-Keys
First seen on t3n.de Jump to article: t3n.de/news/forscher-chatgpt-windows-keys-1696521/
-
UK government signs partnership with OpenAI
Deal sees firm behind ChatGPT collaborate with government on AI security research to explore investment opportunities First seen on computerweekly.com Jump to article: www.computerweekly.com/news/366627871/UK-government-signs-partnership-with-OpenAI
-
Interview: How OpenAI is making ChatGPT public and private sector-ready
We speak to OpenAI’s solution engineering lead, Matt Weaver, about enterprise adoption and making ChatGPT secure First seen on computerweekly.com Jump to article: www.computerweekly.com/news/366627781/Interview-How-OpenAI-is-making-ChatGPT-public-and-private-sector-ready
-
OpenAI stellt ChatGPT Agent vor; Sam Altman warnt vor Risiken
OpenAI hat zum 17. Juli 2025 seinen “ChatGPT-Agenten” veröffentlicht. In einem längeren Post auf X greift OpenAI-Chef Sam Altman dies auf. Gleichzeitig warnt Altman vor potentiell großen Risiken beim Einsatz dieses ChatGPT-Agenten. Es sei alles immer noch experimentell. AI-Agenten sind … First seen on borncity.com Jump to article: www.borncity.com/blog/2025/07/22/openai-stellt-chatgpt-agent-vor-sam-altman-warnt-vor-risiken/
-
ChatGPT’s GPTreasoning-alpha model spotted ahead of launch
Tags: chatgptGPT-5 might be just a few days or weeks away, as we’ve spotted references to a new model called gpt-5-reasoning-alpha-2025-07-13. First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/artificial-intelligence/chatgpts-gpt-5-reasoning-alpha-model-spotted-ahead-of-launch/
-
New ChatGPT o3-alpha model hints at coding upgrade
ChatGPT’s o3 is OpenAI’s best model to date because it features reasoning, and it might get even better in the next update. First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/artificial-intelligence/new-chatgpt-o3-alpha-model-hints-at-coding-upgrade/
-
Ich gebe auf: Wie ein Forscher ChatGPT mit einem simplen Trick dazu bringt, Windows-Keys herauszurücken
First seen on t3n.de Jump to article: t3n.de/news/forscher-knackt-chatgpt-windows-key-mit-trick-1696521/
-
OpenAI’s image model gets built-in style feature on ChatGPT
OpenAI’s image gen model, which is available via ChatGPT for free, now lets you easily create AI images even if you’re not familiar with trends or prompt engineering. First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/artificial-intelligence/openais-image-model-gets-built-in-style-feature-on-chatgpt/
-
OpenAI’s ChatGPT-powered browser is codenamed ‘Aura’
OpenAI is following Perplexity and is working on its own AI-powered browser codenamed “Aura.” First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/artificial-intelligence/openais-chatgpt-powered-browser-is-codenamed-aura/
-
Google’s Gemini refuses to play Chess against the mighty Atari 2600 after realizing it can’t match ancient console
Warned that ChatGPT and Copilot had already lost, it stopped boasting and packed up its pawns First seen on theregister.com Jump to article: www.theregister.com/2025/07/14/atari_chess_vs_gemini/
-
Putting AI-assisted ‘vibe hacking’ to the test
Tags: access, ai, attack, chatgpt, cyber, cybercrime, cybersecurity, data-breach, defense, exploit, hacking, least-privilege, LLM, network, open-source, strategy, threat, tool, vulnerability, zero-trustUnderwhelming results: For each LLM test, the researchers repeated each task prompt five times to account for variability in responses. For exploit development tasks, models that failed the first task were not allowed to progress to the second, more complex one. The team tested 16 open-source models from Hugging Face that claimed to have been…
-
Ich gebe auf: Wie ein Forscher ChatGPT mit einem einfachen Trick dazu bringt, Windows-Keys herauszurücken
First seen on t3n.de Jump to article: t3n.de/news/chatgpt-jailbreak-trick-ki-windows-lizenzschluessel-1696521/
-
Researchers Trick ChatGPT into Leaking Windows Product Keys
Security researchers have successfully demonstrated a sophisticated method to bypass ChatGPT’s protective guardrails, tricking the AI into revealing legitimate Windows product keys through what appears to be a harmless guessing game. This discovery highlights critical vulnerabilities in AI safety mechanisms and raises concerns about the potential for more widespread exploitation of language models. The Gaming…