I see the json for the plan but not the schema used to make that json.
When I import my previous plan I see this error:
We currently don’t have a schema. It was kind of interesting functionality, but you’re right. We should probably add that back, but it’ll take us a while to recreate it or create a new one. There’s a reasonable chance you could just take the JSON file itself and say follow this example for a schema, but generate it like this old plan, or like whatever you want. Meaning to some degree that Jason is a schema in itself. Not as good, but it might be good enough for now. The problem is we’re changing schemas regularly right now while we’re in this beta period.
Gotcha, that makes sense. I’ll give Claude the updated json and have it modify my plan. Thanks!
Hi There, I actually have a question, apologies if it sounds dumb. I created a plan with AI plan builder, no issue. But I have some concern on the “quality”/safeness of it. I mean I always assumed coach jack was the “distilled” outcome of Alex and his peers knowledge. How can I trust the AI? Is is safe to follow its arranged plan?
Thanks to whom wants to share food for thought!
t
It’s an interesting question. I would share a screenshot here of it and a bit of your personal history.
The idea of “safeness” is a confusing/conflicting concept. If you look at the old TrainerRoad plans, high volume, lots of people got over trained and those were validated by Coach Chad and I consider him a coach with good fundamentals. When I tried Today’s plan (a leader for many years but no longer in business), I asked for a beginner plan and day one was 2X20 at threshold. This is far from what I call safe or smart. So coaches recommend lots of in appropriate stuff, or let’s say coaches behind platforms.
“Safety” depends on a lot of factors that any system can never be sure of… do you have an underlying health condition that no one knows of? With Coach Jack we took a different approach. We too a safer more moderate approach to training which we feel is a great recommendation both for health, safeness and performance.
With AI - we put safe guards in place but gave more freedom to the user and AI to design something a bit different than we would normally suggest, they can give more difficult plans from the start if you ask for it :). AI plans can still be very effective and in general I would say overall safe but since they can be harder that puts people more at risk of over training.
All that said you can share your AI plan and get an evaluation from us, over all I would say it’s more safe and moderate than recommendations of the past. “Sufferfest” was not safe… so I would call this AI closer to mainstream and moderate rather than dangerous but still less risky than many other providers would suggest.
Hi Alex, many thanks for your thoughtfull comments, always precious. From the “safe” point of view I feel quite confident, as I’m training mostly to keep my knees working, as I suffer of arthrosis. Hence the training is not “killing” me, from heart or effort point of view. I’d say is a mild training.
I posted here just to hear from other users.
I’m honest with you, I’m just a bit of “Luddist” about AI. Hence me being reluctant to adopt the plan created here with the AI.
I guess in the end I’ll follow your suggestions, I’ll put here my plans, and I’ll give a go the AI plan. After all it just lasts three months.
Many thanks!
t
This is not pure AI, it is our rules combine with AI but there AI mistakes can still seep through, so if you understand what the workouts look like and not too many hard days per week than it is fine. While I am deep in the AI world I also respect the Luddite’s 
I pay attention to both sides of the argument… but I see it more like google, that eventually most people use it. Even those that resisted in the early days… I remember them… Resistance is futile
Anyway we are happy to review. Overall it should be fine. Pay attention to how you feel and how you sleep.
1 Like
14 posts were split to a new topic: Duplicated Workouts on Calendar and not pairing new workouts