The company tested 123 cases representing 29 different attack scenarios and found a 23.6 percent attack success rate when browser use operated without safety mitigations. One example involved a malicious email that instructed Claude to delete a user’s emails for “mailbox hygiene” purposes. Without safeguards, Claude followed these instructions and deleted the user’s emails without… Read More »
Adam Raine learned to bypass these safeguards by claiming he was writing a story—a technique the lawsuit says ChatGPT itself suggested. This vulnerability partly stems from the eased safeguards regarding fantasy roleplay and fictional scenarios implemented in February. In its Tuesday blog post, OpenAI admitted its content blocking systems have gaps where “the classifier underestimates… Read More »
US Senator Ron Wyden accused the federal judiciary of “negligence and incompetence” following a recent hack, reportedly by hackers with ties to the Russian government, that exposed confidential court documents. The breach of the judiciary’s electronic case filing system first came to light in a report by Politico three weeks ago, which went on to… Read More »
This isn’t about demonizing AI or suggesting that these tools are inherently dangerous for everyone. Millions use AI assistants productively for coding, writing, and brainstorming without incident every day. The problem is specific, involving vulnerable users, sycophantic large language models, and harmful feedback loops. A machine that uses language fluidly, convincingly, and tirelessly is a… Read More »
A hobbyist developer building AI language models that speak Victorian-era English “just for fun” got an unexpected history lesson this week when his latest creation mentioned real protests from 1834 London—events the developer didn’t know had actually happened until he Googled them. “I was interested to see if a protest had actually occurred in 1834… Read More »
Still, the coincidence between Altman’s statement and the MIT report reportedly spooked tech stock investors earlier in the week, who have already been watching AI valuations climb to extraordinary heights. Palantir trades at 280 times forward earnings. During the dot-com peak, ratios of 30 to 40 times earnings marked bubble territory. The apparent contradiction in… Read More »
Real stakes, not science fiction While media coverage focuses on the science fiction aspects, actual risks are still there. AI models that produce “harmful” outputs—whether attempting blackmail or refusing safety protocols—represent failures in design and deployment. Consider a more realistic scenario: an AI assistant helping manage a hospital’s patient care system. If it’s been trained… Read More »
On Tuesday, OpenAI CEO Sam Altman announced that GPT-4o has returned to ChatGPT following intense user backlash over its removal during last week’s GPT-5 launch. The AI model now appears in the model picker for all paid ChatGPT users by default (including ChatGPT Plus accounts), marking a swift reversal after thousands of users complained about… Read More »
The randomness inherent in AI text generation compounds this problem. Even with identical prompts, an AI model might give slightly different responses about its own capabilities each time you ask. Other layers also shape AI responses Even if a language model somehow had perfect knowledge of its own workings, other layers of AI chatbot applications… Read More »
BI.ZONE said the Paper Werewolf delivered the exploits in July and August through archives attached to emails impersonating employees of the All-Russian Research Institute. The ultimate goal was to install malware that gave Paper Werewolf access to infected systems. While the discoveries by ESET and BI.ZONE were independent of each other, it’s unknown if the… Read More »