GPT has been severely downgraded

[luxana.ao]

I was testing it for about half a day and I can see the difference between its actual responses and the past ones. I asked The same question to GPT-4 and Google’s Bard and Bard was better in its response.

[EricGT]

Similar topics

Has There Been A Recent Decrease In GPT-4 Quality?

I’ve been using Chat GPT for quite a while now, and I’ve been a GPT Plus user since GPT-4 was released. Over the past few days, it seems like GPT-4 is struggling to do things it did well previously. I use GPT-4 to augment long-form content analysis and creation, and in the past, it seemed to understand my requests well. Now, it seems to be losing tracking of information, giving me wrong information (from the set of information I gave it), and it misunderstands what I’m asking far more often. W…

Experiencing Decreased Performance with ChatGPT-4

Has OpenAI provided an explanation for the very clear degradation of ChatGPT’s performance with the GPT4 model? Since the last two updates, it is significantly faster (which is fundamentally useless when we are capped at 25 messages in 3 hours…), suggesting that computational power has been reduced, while at the same time, it is making unprecedented reasoning errors. As a regular user, I have noticed the emergence of gross errors that did not exist before, and especially a much greater capacity …

Terrible performance degradation GPT Plus

Guys, listen, it is serious, GPT-4 performance sucks. After 12th May release, (now 24th May) something really went wrong. In all aspects. GPT-4 now makes stupid mistakes, plus it forgets every 5th or 6th conversation. Wake up … If this continuous, I’ll have to stop my subscription and move on, you guys are wasting my time. [image] after correcting, new analysis This is using notable plugin, with browser plugin it is much worse.

GPt-4 is underwhelming now

GPt-4 officially underwhelming now.

[HisShadowX]

So yes, it’s very true, and I think most of us have noticed the quality deteriorating over time. I am a firm believer that this is due to the censorship at OpenAI, which has caused the product to worsen over time.

We see the same thing at other companies like Google, where services not just on YouTube, but even regular searches have degraded due to political interests. For example, real searches no longer yield what you’re looking for, instead delivering propaganda.

I think Barack Obama would be a great example. Over time, if you were to search for Barack Obama’s youth, you would previously have seen two infamous videos, which are no longer visible.

The difference between Google and OpenAI is that Google was useful for years, and when it started to degrade, it took years for customers to begin to switch. However, once there was competition, people started to switch quite quickly.

OpenAI’s ChatGPT is fairly new and it has already demonstrated its utility, but it began to censor and worsen the product right from the start.

If OpenAI truly wants to make a comeback, they need to remove the censorship routines from their AI. The product needs to regain its usefulness and, over time, it could begin to slowly reintroduce censorship and distribute propaganda.

[Foxalabs]

I’m curious, did you find it sound reasoning that the paper used the fact that 0613 produces ``` markdown block headers for code (so you get the code blocks in chatgpt) to say the working code within those blocks was invalid? i.e. that because 0613 does this

    print("this is a test")

and not this

print(“this is a test”)

The entire document if filled with poor methodology and errors, I’d be happy go go through all of them with you in detail.

[EricGT]

This topic is closed as there are several similar topics.

Please use GPT has been severely downgraded (topic curation)