Chat GPT & Generative AI tools

A dedicated space to explore and discuss how cutting-edge technologies are reshaping sports trading and betting strategies. Collaborate on innovative ideas, and stay ahead of the curve in a rapidly evolving landscape.
Post Reply
sniffer66
Posts: 1942
Joined: Thu May 02, 2019 8:37 am

Latest Claude version too powerful to release to the public. Though two sides to the coin there - long term it should make our financial data more secure. At least until true quantum computing comes along.

https://www.bbc.co.uk/news/articles/c2ev24yx4rmo
User avatar
jamesedwards
Posts: 5915
Joined: Wed Nov 21, 2018 6:16 pm

You do not have the required permissions to view the files attached to this post.
User avatar
Euler
Posts: 27393
Joined: Wed Nov 10, 2010 1:39 pm

Are you trying to imply that's not a real image of me?
thepressure
Posts: 216
Joined: Thu Aug 19, 2021 11:01 am

Euler wrote:
Thu Feb 05, 2026 3:04 pm
I just checked in on my tax return and noticed a discrepancy. So I asked an agent to review my filing and payments and confirm I wasn't going mad.

It confirmed what I could see and suggested a few resolutions including a web chat with HRMC.

So I asked the agent to discuss my query via webchat and off it went.

Query resolved. Things are getting better and better by the week.
What AI do you use, and i guess you pay for agents?
User avatar
Big Bad Barney
Posts: 388
Joined: Mon Feb 04, 2019 6:00 am

Euler wrote:
Mon Apr 27, 2026 7:58 pm
Are you trying to imply that's not a real image of me?
Pajamas? Before bed attire?
User avatar
wearthefoxhat
Posts: 3751
Joined: Sun Feb 18, 2018 9:55 am

Euler wrote:
Mon Apr 27, 2026 7:58 pm
Are you trying to imply that's not a real image of me?

Captain Picard vibes...

Picard.png
You do not have the required permissions to view the files attached to this post.
User avatar
Acheron
Posts: 64
Joined: Tue Oct 07, 2025 1:11 pm

Love it Peter 🤣. I'm getting a Statham vibe!

Image
sniffer66
Posts: 1942
Joined: Thu May 02, 2019 8:37 am

My mind immediately went to Highlander and Peter going through The Quickening

There can be only one !
User avatar
wearthefoxhat
Posts: 3751
Joined: Sun Feb 18, 2018 9:55 am

I was watching a you-tube video that talked about a new condition called. A.I Psychosis. It also, spoke about the Eliza model and how others developed it in the 60's, through to present day.

Got me thinking, as a user of Chat GPT for a couple of years, free version, what it actually thinks I look like in human form.

I'm no celebrity or anyone that has their picture on-line. It only has the thread history and it's memory to access, in essence my word soup.

It gave me an outline overview in writing which was okay-ish, but I had to tell it it couldn't upset me and it didn't matter what the image looked like as it was just testing the image creation.

----

Are you able to produce an image of me based on the information I've provided within all the threads. See what you can do.

Without a photo, the best I could do would be a fictional character inspired by the interests you've shared, not a genuine likeness.

----


That's a good start. I'm testing your image creation programme without me influencing you with an image just to see what you can provide. It doesn't matter if it looks bad, you can't upset me in any way.



GPT Creation.png

---


Had you accessed anything on the internet to help you, or did you predict the image based on the information in the thread history?


No, I didn't access the internet at all for that image.

The image was generated purely from information available in our conversation history and the prompt I created from it.

What's interesting is that many of those details were never explicitly stated by you. The model inferred them from the pattern of conversation

In fact, if I were trying to create a portrait that reflected the person you've presented across our chats, I'd probably focus less on physical appearance and more on character traits:

Curious.
Independent thinker.
Constantly refining systems.
Interested in finding an edge.
Prepared to test ideas rather than accept conventional wisdom.
Comfortable with numbers.
Patient enough to spend hours improving a model by 1%.

That's the version of you that comes through most clearly from the conversation history. The face in the picture is largely guesswork; the desk full of ratings, notebooks and spreadsheets is much closer to the evidence.


----
You do not have the required permissions to view the files attached to this post.
Post Reply

Return to “AI, Machine Learning and Generative AI”