The Incentive to Deceive

Rob Miles is one of the better explainers of AI on YouTube, he's detailed, he rarely holds back on calling out elephants, and, importantly for broadcast media, he's personable. He's also has a long, in YouTube terms, track record of covering Alignment issues. As a PhD student he's particularly adept at explaining the complexities of Alignment issues. In this video he gives a fine explanation of the reward training in LLM's both implying and stating the issues that ensue from such training approaches, including the policies to please humans, and the utility of such models to deceive.

Two parts near the end of the video caught my attention:

'This is potentially fairly dangerous, there are certain type of goals that are instrumentally valuable for a wide range of different terminal goals, in the sense that, you can't get what you want if you're turned off, you can't get what you want if you're modified, you probably want to gain power and influence."

'Reinforcement Learning From Human Feedback, is a powerful Alignment technique, in a way, but it does not solve the problem...extremely powerful systems trained in this way, I don't think they'd be safe.'

Comments

OpenAI's NSA Appointment Raises Alarming Surveillance Concerns

The recent appointment of General Paul Nakasone, former head of the National Security Agency (NSA), to OpenAI's board of directors has sparked widespread outrage and concern among privacy advocates and tech enthusiasts alike. Nakasone, who led the NSA from 2018 to 2023, will join OpenAI's Safety and Security Committee, tasked with enhancing AI's role in cybersecurity. However, this move has raised significant red flags, particularly given the NSA's history of mass surveillance and data collection without warrants. Critics, including Edward Snowden, have voiced their concerns that OpenAI's AI capabilities could be leveraged to strengthen the NSA's snooping network, further eroding individual privacy. Snowden has gone so far as to label the appointment a "willful, calculated betrayal of the rights of every person on Earth." The tech community is rightly alarmed, with many drawing parallels to dystopian fiction. The move has also raised questions about ...

What is happening inside of the black box?

Neel Nanda is involved in Mechanistic Interpretability research at DeepMind, formerly of AnthropicAI, what's fascinating about the research conducted by Nanda is he gets to peer into the Black Box to figure out how different types of AI models work. Anyone concerned with AI should understand how important this is. In this video Nanda discusses some of his findings, including 'induction heads', which turn out to have some vital properties. Induction heads are a type of attention head that allows a language model to learn long-range dependencies in text. They do this by using a simple algorithm to complete token sequences like [A][B] ... [A] -> [B]. For example, if a model is given the sequence "The cat sat on the mat," it can use induction heads to predict that the word "mat" will be followed by the word "the". Induction heads were first discovered in 2022 by a team of researchers at OpenAI. They found that induction heads were present in ...

Prompt Engineering: Expert Tips for a variety of Platforms

Prompt engineering has become a crucial aspect of harnessing the full potential of AI language models. Both Google and Anthropic have recently released comprehensive guides to help users optimise their prompts for better interactions with their AI tools. What follows is a quick overview of tips drawn from these documents. And to think just a year ago there were countless YouTube videos that were promoting 'Prompt Engineering' as a job that could earn megabucks... The main providers of these 'chatbots' will hopefully get rid of this problem, soon. Currently their interfaces are akin to 1970's command lines, we've seen a regression in UI. Constructing complex prompts should be relegated to Linux lovers. Just a word of caution, even excellent prompts don't stop LLM 'hallucinations'. They can be mitigated against by supplementing a LLM with a RAG, and perhaps by 'Memory Tuning ' as suggested by Lamini (I've not tested this approach yet). ...

Charting the emergence of AGI?

Search This Blog