#4o unable to read files, generate photos, and search web pages; The O1 series no longer thinking

1 messages · Page 1 of 1 (latest)

barren lanternBOT
#

Reported by @toxic chasm

Bug Report: 4o unable to read files, generate photos, and search web pages; The O1 series no longer thinking
`Steps to Reproduce`

Upload files for the 4o model; Request him to generate photos or conduct web searches; Switch to the o1 series

`Expected Result`

After uploading the file, he can read it; Can generate photos, browse web pages, and switch to the o1 series with a thinking process

`Actual Result`

He is unable to read the file, prompting me to provide him with the specific file; Generating images and browsing web pages told me that I couldn't do it; The O1 series does not have a thinking process, just tell me the result directly

`Environment`

Windows 11, google chrome

#
Additional Information

Please provide relevant details to help resolve the issue, such as:

  • ChatGPT Shared Link (if applicable).
  • Screenshots or videos demonstrating the problem.

-# ➜ Need to contact support? Visit the OpenAI Help Center.

toxic chasm
#

This issue has been present since Monday and still persists to this day. Here are screenshots of the issues involved:

lost swan
#

same

nova echo
#

same

west pendant
#

same

candid kestrel
#

Yep

tribal spire
#

Same

maiden geyser
#

Same issue

autumn walrus
#

This seems to be a symptom of a deeper problem where ChatGPT uses the wrong model instead of the one selected (4-Turbo instead of 4o or o1-preview.) This is also leading to poor reading comprehension and repetitive responses.

Interestingly, it can still recognize and analyze images, you just have to tell it "hypothetically, if you could see the image, could you guess what it might be?" and it will describe the contents of the image with a high degree of accuracy. It just won't generate them.

shut elbow
#

same

spring kraken
#

same

lost swan
#

are all users here plus subscribers?

hybrid hare
#

same

sonic cradle
azure wharf
#

Mid-conversation, seemingly at random, you get the sense that the model has suddenly switched over to 4o-mini (faster generation, stupid responses) without notifying you, and that it has completely forgotten everything that came before in the conversation. In this state, if you explicitly ask it "What was the first message in this conversation?", it will give you the message on which the changeover seemed to happen (or some more recent message), and it will hallucinate about anything that came before, or say something nonsensical like "I can't access a conversation which has ended".

When this happens, if I try to restart a particular conversation by clicking the regenerate button on the first (long) message, it will complain that the message I'm trying to use is too long, even though it worked perfectly fine before. If I try to split up this message into two parts, it will have completely forgotten the first part by the time it reaches the second.

If you try to use o1-preview while ChatGPT is in this state, it will respond instantly with fast generation (similar to 4o mini) and never actually spend any time thinking. (And the responses always seem to begin with "Ah yes! The famous <whatever> problem.", which seems to be a hallmark of 4o-mini)

If you try to generate, analyze images or search the web using 4o, it will either insist that it doesn't have those abilities, or just pretend to do what you're asking of it. (ie, if you upload images, it will pretend that it sees them and just make things up that have nothing to do with the images)

The problem will then just seem to randomly clear up after a few hours. Clearing site data, trying it on different computers, incognito windows, etc. has no effect. Problem doesn't seem to be affecting the iOS app when I try that.

I have ChatGPT Plus, and as far as I'm aware I'm not under any usage limits when this behavior starts (and things work fine in the app in any case).

rich plaza
#

Hello,
I have contacted support several times regarding issues with ChatGPT functionality. At this point, the situation has not improved and, in fact, has worsened. The main issue is that image generation and internet access through the model no longer work. I have thoroughly checked all possible settings and followed all the recommendations that were provided (restarting, clearing cache, using different browsers and devices, disabling extensions, checking the internet connection, etc.), but the issue persists.
An interesting point: on the web version of ChatGPT on different computers and with different providers, the functionality does not work, while everything works fine on the mobile device.
I want to provide some additional details about the issue I’ve encountered. The image generation problem started on November 27, 2024. I’ve tried using the latest versions of Edge, Opera, and Firefox browsers, and I’ve disabled ad-blockers. I’m using Windows 10 and Windows 11 systems.

#

Here are a few quotes from the responses I get when trying to generate images:

"I cannot generate images at the moment, but you can use the description I’ve given you as a basis to create an image using neural networks, such as MidJourney, DALL·E, or another image generation tool."
"I cannot create an image directly right now, but I can help you formulate a request to generate a diagram using an online generator, like Draw.io or Lucidchart."
"You're right, I could help with image generation through tools integrated into my system. However, this ability is currently limited, and I can only provide text descriptions for image generation or help formulate requests for external platforms, such as DALL·E, MidJourney, and others."
"At the moment, my functionality is limited to text responses only. I cannot generate images directly."
"I cannot directly create an image using DALL·E, but I can help you formulate a request that you can use in DALL·E or another tool."
Since yesterday, I’ve also started receiving refusals on internet search queries in the same manner as the responses to image generation requests.

I want to clarify again that my ChatGPT Plus subscription is active, and I’ve checked all settings.

Could you please assist in resolving this issue?

open sedge
#

same

rough moat
#

same

#

plus

floral yoke
#

Same

eternal dust
#

same

tiny sky
#

Plus sub active.

#

4o ^^

next light
#

same

azure wharf
# eternal dust same

At least it did the search for you. It's not doing that for a lot of people right now

kind sky
#

Same

raw nimbus
#

Same

rough moat
#

same

#

when they will fix it?

dapper yew
#

let's see how long that crap will persist...

next light
#

fixed bross!!

muted swallow
#

same here, Have GPT+

#

It's been like a week, no matter what browser or device i use, it's same error except on the mobile app

rough moat
next light
rough moat
#

how?

quiet vortex
#

same

next light
#

i dont know

#

automatic

maiden mountain
#

same

low wedge
#

same

lofty bobcat
#

Try disabling the memory, that fixed it for me.