This thinly veiled advertisement claims it's a waste of time to understand the tradeoffs in the models you're using, and you should instead pay them to make those decisions for you. No thank you.
Pretty basic advertisement for a content advertising op. Model picking is bad UX because, so we'll automate it for you, smash that call-to-action button now.
I agree in principle, but I feel like we're not "there" yet, especially for tools like Copilot that allow models from different vendors.
I do expect to see a unified model interface for regular ChatGPT and similar tools this year (especially since sama has been talking about it), but for more technical tools that have various models I'd like to keep the choice overall, at least for right now.
That is a commonly held UX belief - "if the user has made a choice, we have failed".
It's also a fundamentally untrue belief. It discards the fact that different tools have different usages, and it fundamentally says that users do not care to exercise taste, either. (Because "which model" is also a taste function)
The lens I wish UX would take is "how can we make this concept more useful/lower friction" instead of dreaming of annihilating it. That's fundamentally the DWIM dream, and it never works.
Shakahs|1 year ago
perching_aix|1 year ago
headcanon|1 year ago
I do expect to see a unified model interface for regular ChatGPT and similar tools this year (especially since sama has been talking about it), but for more technical tools that have various models I'd like to keep the choice overall, at least for right now.
drcongo|1 year ago
groby_b|1 year ago
It's also a fundamentally untrue belief. It discards the fact that different tools have different usages, and it fundamentally says that users do not care to exercise taste, either. (Because "which model" is also a taste function)
The lens I wish UX would take is "how can we make this concept more useful/lower friction" instead of dreaming of annihilating it. That's fundamentally the DWIM dream, and it never works.