UDHR and Alignment

The Universal Declaration of Human Rights (UDHR) is a document that sets out fundamental human rights to be universally protected. Ideally any alignment of AI should use this as the basis for what human values are. The document was written in 1948 as a response to the atrocities of the Second World War. They remain the clearest expression of human values I know. They have failed though in practice, as I can certainly think of many examples where the post war governments, in countries like the UK, have breached most of the 30 articles stated.

If governments can't or won't follow and uphold 30 basic principles for human values, why is there an expectation that AI can or will be able to?

Cansu Canca considered this issue in a post from 2019 "AI & Global Governance: Human Rights and AI Ethics – Why Ethics Cannot be Replaced by the UDHR" Canca states that 'When we dive deep, the UDHR is simply unable to guide us on those questions. Solving such challenges is the job of ethical reasoning.'

The conclusion of: 'I do not mean to say that the UDHR is not of any use in the discussion of ethical tech. Its clarity, legacy, and wide acceptance makes the UDHR a good tool to use to start the exploration on what might be problematic about any given AI system or practices in developing these systems. However, if the aim is not just to identify the problem but also to solve it, then the UDHR is simply inadequate to do so. Here, I invite you to engage in ethics.'

Comments

The Whispers in the Machine: Why Prompt Injection Remains a Persistent Threat to LLMs

Large Language Models (LLMs) are rapidly transforming how we interact with technology, offering incredible potential for tasks ranging from content creation to complex analysis. However, as these powerful tools become more integrated into our lives, so too do the novel security challenges they present. Among these, prompt injection attacks stand out as a particularly persistent and evolving threat. These attacks, as one recent paper (Safety at Scale: A Comprehensive Survey of Large Model Safety https://arxiv.org/abs/2502.05206) highlights, involve subtly manipulating LLMs to deviate from their intended purpose, and the methods are becoming increasingly sophisticated. At its core, a prompt injection attack involves embedding a malicious instruction within an otherwise normal request, tricking the LLM into producing unintended – and potentially harmful – outputs. Think of it as slipping a secret, contradictory instruction into a seemingly harmless conversation. What makes prompt inj...

Can We Build a Safe Superintelligence? Safe Superintelligence Inc. Raises Intriguing Questions

Safe Superintelligence Inc . (SSI) has burst onto the scene with a bold mission: to create the world's first safe superintelligence (SSI). Their (Ilya Sutskever, Daniel Gross, Daniel Levy) ambition is undeniable, but before we all sign up to join their "cracked team," let's delve deeper into the potential issues with their approach. One of the most critical questions is defining "safe" superintelligence. What values would guide this powerful AI? How can we ensure it aligns with the complex and often contradictory desires of humanity? After all, "safe" for one person might mean environmental protection, while another might prioritise economic growth, even if it harms the environment. Finding universal values that a superintelligence could adhere to is a significant hurdle that SSI hasn't fully addressed. Another potential pitfall lies in SSI's desire to rapidly advance capabilities while prioritising safety. Imagine a Formula One car wi...

AI Agents and the Latest Silicon Valley Hype

In what appears to be yet another grandiose proclamation from the tech industry, Google has released a whitepaper extolling the virtues of what they're calling "Generative AI agents". (https://www.aibase.com/news/14498) Whilst the basic premise—distinguishing between AI models and agents—holds water, one must approach these sweeping claims with considerable caution. Let's begin with the fundamentals. Yes, AI models like Large Language Models do indeed process information and generate outputs. That much isn't controversial. However, the leap from these essentially sophisticated pattern-matching systems to autonomous "agents" requires rather more scrutiny than the tech evangelists would have us believe. The whitepaper's architectural approaches—with their rather grandiose names like "ReAct" and "Tree of Thought"—sound remarkably like repackaged versions of long-standing computer science concepts, dressed up in fashionable AI clot...

Charting the emergence of AGI?

Search This Blog