OpenAI Always one step fwd in the scam level
retained text
With the 5 hour limit on Plus accounts they basically pushed users into using the web version The web version already had this compute limitation where you would suddenly get the Our system needs to check message and the only workaround was to copy everything done so far into a txt file stop the chat paste it back and explain that it had bugged after doing this and that Now they somehow made it even worse by completely removing the reasoning display on that purpose so instead of knowing the answer was paused server side the chat just stays stuck with no message and no indication of what is happening This is by far the worst UX I have experienced with any application in more than 30 years Trying to fight your own customers is not the solution If the limits were not this aggressive people would not be mass buying 5 dollar shared account access but that is exactly what is happening now because there is NO upside anymore to having proper Codex access You can barely work for an hour with Sol Light on Codex before running out of credits You should seriously start using your own product at full capacity to provide a heaven env to your your customers instead of continuing this endless fight against them Making the product harder to use is not going to stop people from finding workarounds it is only going to make paying customers more frustrated
document fetched 2026-09-26 00:28 · canonical https://reddit.com/r/OpenAI/comments/1wpsexi/openai_always_one_step_fwd_in_the_scam_level