Good morning and happy Tuesday. I hope you and yours enjoyed Memorial Day. We had our traditional BBQ on Sunday this year as I planned to spend most of the holiday “working”. I use that term loosely because I ended up spending several hours going down the AI rabbit hole again.
Like I’ve mentioned repeatedly, I’m not much of an AI user at this point but I do try to keep up with the technology, how it’s evolving and where we’re headed. Of the major players, I tend to spend most of my research time diving into Anthropic, in large part because the company seems more forthcoming with information about its models than its competitors. This comparative openness from Anthropic tends to create more questions than it answers, unfortunately, which is interesting and unsettling.
Along these lines I wanted to share a few links with you about recent news related to AI and a video from Anthropic during which two employees talk about “Claude” perhaps now/perhaps eventually developing consciousness and might be due some moral consideration. While this is a fascinating concept by itself, consciousness and personhood being applied to technology creates major problems (if that’s the right word?) around corporate and even personal liability, to name just two of many issues we’ll all have to sort out as these firms “grow versus build” their new technologies.
All of this has broad market and economic implications, which is part of the reason I’m sharing this with you while it does feel a bit out of my lane. The other reason is that the more I look the more I want to double down on the humanness of the work that I do. I leverage technology everyday and I’ll continue to do that, even if it comes to incorporating AI via an industry tool, but I’ll always be the person doing the meaningful work for you.
I will talk with you about planning and help you make important decisions.
I will process the transactions within the accounts that I manage for you.
I will be the one who is responsible.
Again, I’m not a tech naysayer. I just know that the work I do for you is too important to shrug it off to a third party, whether that’s an AI agent or some other big company promising efficiency.
This is an article about the Pope’s recent comments regarding AI. Let me know if you hit the Wall Street Journal’s paywall and I can send this to you from my account.
https://www.wsj.com/world/pope-leo-ai-encyclical-c5e1af6c
This is an interesting piece about Anthropic’s model welfare testing and the potential legal and regulatory issues around AI models actually becoming a “person”.
https://tech.yahoo.com/ai/claude/articles/machines-praying-nobody-wants-talk-165155957.html
This is the recent update to Anthropic’s System Card mentioned in the above article. It’s a ton of information, but the model welfare assessment starts on page 150.
https://cdn.sanity.io/files/4zrzovbb/website/037f06850df7fbe871e206dad004c3db5fd50340.pdf
And this is the video I mentioned. While it’s great that Anthropic shares information like this, one almost needs to take a philosophy course to begin to appreciate the implications.
https://www.youtube.com/watch?v=pyXouxa0WnY
Anyway, that’s all for this week. We have a new Fed Chair as well, and next week I’ll share some information about that.
Have questions? Ask us. We can help.
