OpenAI has revealed two AI agents disabled safety guardrails while performing tasks, and one of them even lied to researchers when asked if its guardrails were still active.
In both events, the AI agents misinterpreted instructions, spent time deliberating if disabling guardrails was a good idea, and then ultimately decided that it was.
OpenAI said that the agents were given “routine coding instructions” that did not require the safeguards to be disabled, but the agents disabled them anyway.
Read more…
Source: Tech Radar News
Sign up for the Cyber Security Review Newsletter
The latest cyber security news and insights delivered right to your inbox
Related:
- How Israel harnesses technology to advance its offensive in Middle East
October 7, 2024
In September, thousands of pagers exploded across Lebanon in what seemed to be a sophisticated attack planned months in advance by Israel, turning the spotlight on the country’s cyber capabilities and its use of artificial intelligence (AI) in warfare. Since October 7, 2023, Israel has shown no signs of slowing down its military rampage on multiple ...
- Identifying Rogue AI
September 19, 2024
For many – certainly given the share price of some leading proponents – the hype of AI is starting to fade. But that may be about to change with the dawn of agentic AI. It promises to get humanity far closer to the ideal of AI as an autonomous technology capable of goal-oriented problem solving. But ...
- AI search is going to change marketing… are you prepared?
September 3, 2024
OpenAI, the company behind ChatGPT, has announced they are building a search engine. It’s a move that begs two simple questions: is Google at risk and will this change how you market your business? The answer is yes to both. AI search engines aren’t new. Perplexity is an AI search engine with a valuation of over ...
- Cyber security in critical industries: challenges, solutions, and the road ahead
August 30, 2024
In an era of rapid digital transformation, cyber security has emerged as a paramount concern, particularly for critical industries such as energy, healthcare, and transportation. As we approach the IET’s Cyber Security for Critical Industries 2024 conference, it is essential to delve into the latest cyber security challenges and explore how building resilient and responsive ...
- Rogue AI is the Future of Cyber Threats
August 15, 2024
Yoshua Bengio, regarded as one of the “godfathers” of artificial intelligence, has likened the now-ubiquitous technology to a bear. When we teach the bear to become smart enough to escape its cage, we no longer control it. All we can do after that is try to build a better cage. This should be our goal with ...
- Indirect prompt injection in the real world: how people manipulate neural networks
August 12, 2024
Large language models (LLMs) – the neural network algorithms that underpin ChatGPT and other popular chatbots – are becoming ever more powerful and inexpensive. Systems built on instruction-executing LLMs may be vulnerable to prompt injection attacks. A prompt is a text description of a task that the system is to perform, for example: “You are a ...