Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Strategies & Market Trends : The Art of Investing -- Ignore unavailable to you. Want to Upgrade?

To: Johnny Canuck who wrote (10606)	11/21/2025 11:59:11 PM
From: Sun Tzu	Read Replies (1) \| Respond to of 10709

No problems at all, but I use the Pro Model, so they may have prioritized me. BTW, I have 3 active patents on hand that I hope to file soon, but since it is late, I decided to dig up one of my long running research projects (the new developments contribute to it). I asked ChatGPT to critique me. This part (a very small part of the feedback) is what I have often worried about...but for the rest of it, it keeps getting warmer. Control authority over the base model You’re implicitly assuming you can: reliably steer the LLM with this external HVS, and that the LLM’s latent space will cooperate. That’s not guaranteed. You’ll need explicit mechanisms (prompting patterns, tool APIs, or smaller steering heads) to make the model obey the worldview rather than ignore it.