> ## Content Index
> Fetch the complete content index at: https://www.cesarrg.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# Trust and alignment
- URL: https://www.cesarrg.com/trust-and-alignment/
- Published: 2023-11-27T20:55:26.000Z
- Updated: 2024-09-07T09:57:13.000Z
- Author: César Rodríguez
- Tags: Design, Technology, AI

#89 - Nov.2023

Trust will continue to be the backbone of good product experiences. This is not different for AI-based products.

AI models, in particular [LLMs](https://www.cesarrg.com/ai-chatgpt/), bring a bigger challenge to serve common affordances. Is the so-called *alignment* problem. The alignment problem in AI is basically reduced to how we create an experience - call it a copilot, agent, AI buddy, etc - to behave in accordance with what we want. 

![](https://storage.ghost.io/c/10/f4/10f4a52f-1b7d-4284-b66d-0fccfd9dc5da/content/images/2023/11/Alignment_.png)

It seems a fair ask for a generative technology with the power of creating content in multiple forms such as text, audio, images, or video. Even more important if we expect the system to behave in accordance with our personal values and [ethics](https://www.cesarrg.com/accelerators-privacy-ethics/). 

The *HHH framing* is a good starting point ([Askell et al. 2021](https://arxiv.org/pdf/2112.00861.pdf?ref=cesarrg.com)). This framing defines alignment in the context of how *helpful, honest, and harmless* can a AI-system be. 

Aligning to our expectations starts with ensuring the system performs the desired task concisely and as effectively as possible (helpful). The task/answer should be accurate or acknowledge the level of [uncertainty](https://www.cesarrg.com/probabilistic-experiences/) (honest). Finally, AI systems should not be offensive or discriminatory, or promote dangerous behaviors (harmless). 

[Alignment is all we need](https://arxiv.org/abs/2205.10268?ref=cesarrg.com). 

PS: Interesting paper of potential solutions on how to solve the alignment problem:

[Personal Universes:A Solution to the Multi-Agent Value Alignment Proble1901.01851.pdf397 KBdownload-circle](https://www.cesarrg.com/content/files/2024/09/1901.01851.pdf "Download")

![](https://storage.ghost.io/c/10/f4/10f4a52f-1b7d-4284-b66d-0fccfd9dc5da/content/images/2024/02/CR_sign_2024-3.png) 

## Get weekly inspiration 💡 

Every week I share quick knowledge bites about design, tech innovation, and creativity. Join others and subscribe for FREE to stay up-to-date with the latest ideas.

Try it 

Email sent! Check your inbox to complete your signup. 

All practical learnings. No spam, ever.