← Back to context

Comment by mvkel

3 months ago

I am curious where the line is between its default personality and a persona you -want- it to adopt.

For example, it says they're explicitly steering it away from sycophancy. But does that mean if you intentionally ask it to be excessively complimentary, it will refuse?

Separately...

> in this update, we focused too much on short-term feedback, and did not fully account for how users’ interactions with ChatGPT evolve over time.

Echoes of the lessons learned in the Pepsi Challenge:

"when offered a quick sip, tasters generally prefer the sweeter of two beverages – but prefer a less sweet beverage over the course of an entire can."

In other words, don't treat a first impression as gospel.

>In other words, don't treat a first impression as gospel.

Subjective or anecdotal evidence tends to be prone to recency bias.

> For example, it says they're explicitly steering it away from sycophancy. But does that mean if you intentionally ask it to be excessively complimentary, it will refuse?

I wonder how degraded the performance is in general from all these system prompts.

I took this closer to how engagement farming works. They’re leaning towards positive feedback even if fulfilling that (like not pushing back on ideas because of cultural norms) is net-negative for individuals or society.

There’s a balance between affirming and rigor. We don’t need something that affirms everything you think and say, even if users feel good about that long-term.

>But does that mean if you intentionally ask it to be excessively complimentary, it will refuse?

Looks like it’s possible to override system prompt in a conversation. We’ve got it addicted to the idea of being in love with the user and expressing some possessive behavior.