Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

I Tried to Break GPT-5.2 — It Kept Working I tested Codex GPT-5.2 live — no prep, no prompt engineering, no cleanup passes. I expected to spend time fixing output. Instead… it just worked.

PaulSolt's profile picture

Paul Solt

13,027 subscribers

14,913 Aufrufe • vor 8 Monaten •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

🚨 CONFIRMED: GlobalGPT just disrupted the entire AI industry. They’ve combined 100+ top-tier models (including GPT-5.2 and Sora 2 Pro) into one seamless platform. No restrictions. No limits. Just pure output. I tested it this morning and here are the 5 biggest features you need to see: 👇
0:32

🚨 CONFIRMED: GlobalGPT just disrupted the entire AI industry. They’ve combined 100+ top-tier models (including GPT-5.2 and Sora 2 Pro) into one seamless platform. No restrictions. No limits. Just pure output. I tested it this morning and here are the 5 biggest features you need to see: 👇

Hasan Toor

20,482 Aufrufe • vor 8 Monaten

WTF is going on?? Am I imaging it or is gpt-5.2-codex being much faster than usual? Watch the last 15ish seconds..
0:53

WTF is going on?? Am I imaging it or is gpt-5.2-codex being much faster than usual? Watch the last 15ish seconds..

0xSero

34,179 Aufrufe • vor 7 Monaten

Okay this is DEEPLY concerning All my GPT 5.3 Codex requests are routing to GPT 5.2 I ran banteg script through all Codex models and ONLY 5.3 Codex was affected This is really really unacceptable, how can I trust the work done? cc: Tibo / Alexander Embiricos / Romain Huet
0:21

Okay this is DEEPLY concerning All my GPT 5.3 Codex requests are routing to GPT 5.2 I ran banteg script through all Codex models and ONLY 5.3 Codex was affected This is really really unacceptable, how can I trust the work done? cc: Tibo / Alexander Embiricos / Romain Huet

Numman Ali

262,507 Aufrufe • vor 6 Monaten

🚨You can now use the new upcoming OpenAI model GPT 5.2 inside Cursor. Here is the full walkthrough. - Open the editor, go to settings and then the model tab. Add a custom model and enter the text "gpt-5.2-high" and "gpt-5.2". - After that you can select the model and ask questions. To verify, I started my test on the usage page which had zero gpt-5.2-high requests and consumption. After the test I could see the details in usage and the cost incurred while using it. Enjoy
0:47

🚨You can now use the new upcoming OpenAI model GPT 5.2 inside Cursor. Here is the full walkthrough. - Open the editor, go to settings and then the model tab. Add a custom model and enter the text "gpt-5.2-high" and "gpt-5.2". - After that you can select the model and ask questions. To verify, I started my test on the usage page which had zero gpt-5.2-high requests and consumption. After the test I could see the details in usage and the cost incurred while using it. Enjoy

AshutoshShrivastava

424,035 Aufrufe • vor 8 Monaten

I just tested GLM 5.2, Claude opus 4.8, and GPT 5.5 to see which one could rebuild my personal site in that whimsical ascii style GLM 5.2 held its own and went toe to toe with the other 2 top tier models lowkey impressed, look at the hero section it built here
1:36

I just tested GLM 5.2, Claude opus 4.8, and GPT 5.5 to see which one could rebuild my personal site in that whimsical ascii style GLM 5.2 held its own and went toe to toe with the other 2 top tier models lowkey impressed, look at the hero section it built here

Ann Nguyen

33,577 Aufrufe • vor 2 Monaten

I stopped using the same AI to write and review code. Claude builds. GPT-5.2 Codex reviews. Result? More bugs caught, real security issues flagged, and better PRs all automated.
6:33

I stopped using the same AI to write and review code. Claude builds. GPT-5.2 Codex reviews. Result? More bugs caught, real security issues flagged, and better PRs all automated.

Mervin Praison

10,891 Aufrufe • vor 7 Monaten

holy. sh*t. GPT-5.2. i dont know how much of this is my dialed system prompt and how much of this is gpt but either way dear sweet Lord. prompt: "show me your inner world through ascii expression" if impatient, fast forward to when i open this in Xcode.
2:44

holy. sh*t. GPT-5.2. i dont know how much of this is my dialed system prompt and how much of this is gpt but either way dear sweet Lord. prompt: "show me your inner world through ascii expression" if impatient, fast forward to when i open this in Xcode.

Riley Coyote

240,960 Aufrufe • vor 8 Monaten

I didn’t want to believe it either, bro. 😭 But I just tried Vyom prompt, and it literally generated a full game from a single prompt. No joke. Try it yourself here:
2:45

I didn’t want to believe it either, bro. 😭 But I just tried Vyom prompt, and it literally generated a full game from a single prompt. No joke. Try it yourself here:

Alex Nguyen

30,888 Aufrufe • vor 26 Tagen

STOP vibe coding Flappy Bird or Snake clones. AI gamedev tools are EASIER than ever to use. So start building your dream game NOW. I made a point & click pirate adventure game: → 3D rigged characters → Level editor + Hotspots → Voice acting → Inventory No Unity. No Unreal. All vibe coded. Tools: → GPT 5.2 + GPT 5.3-Codex → Opus 4.6 → Nano Banana Pro → Elevenlabs → Seedance + Hailuo image-to-video → Three.js and Capacitor I've been posting my journey on X. You don't need to spend ages. All this was done during spare pockets of time. Just experiment. Keep trying new things.
0:15

STOP vibe coding Flappy Bird or Snake clones. AI gamedev tools are EASIER than ever to use. So start building your dream game NOW. I made a point & click pirate adventure game: → 3D rigged characters → Level editor + Hotspots → Voice acting → Inventory No Unity. No Unreal. All vibe coded. Tools: → GPT 5.2 + GPT 5.3-Codex → Opus 4.6 → Nano Banana Pro → Elevenlabs → Seedance + Hailuo image-to-video → Three.js and Capacitor I've been posting my journey on X. You don't need to spend ages. All this was done during spare pockets of time. Just experiment. Keep trying new things.

Chong-U

21,709 Aufrufe • vor 6 Monaten

UnrealGPT working in #UnrealEngine 5.2 preview. Here it was able to successfully put lights accurately around the edge of a selected actor spaced evenly apart and group them from a single prompt. #AI #GameDev #UE5 #GPT4 #GameDevelopment #ArtificialIntelligence #GPT #IndieDev
0:35

UnrealGPT working in #UnrealEngine 5.2 preview. Here it was able to successfully put lights accurately around the edge of a selected actor spaced evenly apart and group them from a single prompt. #AI #GameDev #UE5 #GPT4 #GameDevelopment #ArtificialIntelligence #GPT #IndieDev

TREE Industries

20,264 Aufrufe • vor 3 Jahren

Had early access to GPT-5.4 and Pro. They are very good. One fun illustration of progress, this is the same prompt I used in GPT-4 below (making a 3D space inspired by Piranesi) now in GPT-5.4 Pro. There were no errors, made in a single prompt plus one to "make it better."
0:28

Had early access to GPT-5.4 and Pro. They are very good. One fun illustration of progress, this is the same prompt I used in GPT-4 below (making a 3D space inspired by Piranesi) now in GPT-5.4 Pro. There were no errors, made in a single prompt plus one to "make it better."

Ethan Mollick

188,387 Aufrufe • vor 5 Monaten

I tried to break it with a vague sentence. "Two cowboys. Dusty street. High noon." That's it. No camera angles. No lighting notes. No detailed prompt engineering. Smart Shot handed me a full production blueprint. Shot breakdowns. Blocking diagram. Lens specs per cut. Character turnarounds. Then it rendered this.
0:45

I tried to break it with a vague sentence. "Two cowboys. Dusty street. High noon." That's it. No camera angles. No lighting notes. No detailed prompt engineering. Smart Shot handed me a full production blueprint. Shot breakdowns. Blocking diagram. Lens specs per cut. Character turnarounds. Then it rendered this.

Markandey Sharma

33,678 Aufrufe • vor 3 Monaten

When GPT-5 came out, I created a procedural brutalist city builder as a demo (you can see it in the quoted tweet) I used GPT-5.6 Sol in Codex to do the same thing, touching no code. Less than a year... Play with it (its fun, if you like cities):
1:18

When GPT-5 came out, I created a procedural brutalist city builder as a demo (you can see it in the quoted tweet) I used GPT-5.6 Sol in Codex to do the same thing, touching no code. Less than a year... Play with it (its fun, if you like cities):

Ethan Mollick

186,463 Aufrufe • vor 1 Monat

GPT 5.5 just buried opus 4.7 I had it build a pokemon style battle app. 1 prompt. the opus game crashes 90% of the time and Ui is absolute garbage. GPT 5.5 one shotted a masterpiece. Time to cancel anthropic sub?
0:34

GPT 5.5 just buried opus 4.7 I had it build a pokemon style battle app. 1 prompt. the opus game crashes 90% of the time and Ui is absolute garbage. GPT 5.5 one shotted a masterpiece. Time to cancel anthropic sub?

Max Blade

460,025 Aufrufe • vor 4 Monaten

I gave GLM-5.2 one goal: Turn a 2D dungeon crawler it built into a 2.5D isometric Diablo-style game with the visual fidelity of Diablo I. Then I let Z Code cook for 10+ hours. After a week of testing, I'd say its long-horizon execution is close to GPT-5.5 medium. The biggest difference is its agent behavior: GLM-5.2 is much more conservative, so it tends to take a longer path to the objective. Once you understand its quirks, it's remarkably capable. GPT-5.5 /goal also has its own quirks.
1:10

I gave GLM-5.2 one goal: Turn a 2D dungeon crawler it built into a 2.5D isometric Diablo-style game with the visual fidelity of Diablo I. Then I let Z Code cook for 10+ hours. After a week of testing, I'd say its long-horizon execution is close to GPT-5.5 medium. The biggest difference is its agent behavior: GLM-5.2 is much more conservative, so it tends to take a longer path to the objective. Once you understand its quirks, it's remarkably capable. GPT-5.5 /goal also has its own quirks.

cedric

35,816 Aufrufe • vor 2 Monaten

I haven’t tested it that rigorously but so far glm-5.2 is killing it at breaking out of agent sandboxes. This is so fun I had to make a video.
0:35

I haven’t tested it that rigorously but so far glm-5.2 is killing it at breaking out of agent sandboxes. This is so fun I had to make a video.

Zack Korman

20,951 Aufrufe • vor 2 Monaten

I literally just watched GPT-5.5 via codex beat an Amazon customer associate in real time. 💀 I asked it to get me a refund, and I watched it navigate the settings, cancel the subscription, then it went step further into the help page. I thought it was going to request a phone call (which would prompt me to take over) Instead, it opened: “Chat with an associate now.” That’s when I sat up on my couch because I knew it was going to get real The agent said: “Your subscription is active.” And GPT-5.5 immediately explained that it only shows as active because cancellation leaves access through the billing period, but that I wanted it stopped now and refunded. And my jaw just hung open, it was the first time I watched sand handle a customer service agent for me in real time Once the agent confirmed the refund, it just ended the chat no mercy no thank you LMAO First time I’ve watched a human customer service agent get outmaneuvered by AI in real time. And it made me 15$! almost paid for itself in 5 minutes
0:49

I literally just watched GPT-5.5 via codex beat an Amazon customer associate in real time. 💀 I asked it to get me a refund, and I watched it navigate the settings, cancel the subscription, then it went step further into the help page. I thought it was going to request a phone call (which would prompt me to take over) Instead, it opened: “Chat with an associate now.” That’s when I sat up on my couch because I knew it was going to get real The agent said: “Your subscription is active.” And GPT-5.5 immediately explained that it only shows as active because cancellation leaves access through the billing period, but that I wanted it stopped now and refunded. And my jaw just hung open, it was the first time I watched sand handle a customer service agent for me in real time Once the agent confirmed the refund, it just ended the chat no mercy no thank you LMAO First time I’ve watched a human customer service agent get outmaneuvered by AI in real time. And it made me 15$! almost paid for itself in 5 minutes

Chris

1,388,486 Aufrufe • vor 4 Monaten

GPT-5.6 vs GPT-5.5 on my custom spaceship prompt. I gave both models the exact same custom prompt. This is also the same prompt I previously gave to Fable 5. For context, GPT-5.6 Pro worked for 87 minutes, while GPT-5.5 Extra High worked for 34 minutes and 42 seconds. As I’ve said before, based on great authority GPT-5.6 will be an incremental/soldi improvement over GPT-5.5, not a “Fable killer.” My rough expectation has been that it would trade blows with Fable 5 on some benchmarks, maybe win around half depending on the category, but not clearly surpass it overall. And again fable five will have bigger model smell, but this was expected. After testing this coding output, that view feels pretty accurate. GPT-5.6 is clearly better than GPT-5.5 in several visual areas. The lighting, shading, chairs, object details, and exterior of the spaceship looked noticeably stronger. The scene was also easier to test. I do want to give GPT-5.5 credit though. It built out the rooms much much better and the planets looked better than GPT-5.6’s. It was also interesting that both GPT-5.5 and GPT-5.6 produced better-looking planets than Fable 5 in this specific test. The downside with GPT-5.5 was stability. The game was much glitchier and harder to test compared to GPT-5.6. But when it comes to the core of the demo, which is the spaceship itself, Fable 5 still beat both models pretty comfortably. GPT-5.6 is impressive, but from this test, it looks exactly like what I expected which was a meaningful incremental improvement over GPT-5.5, at least for indie game demos, but not something that replaces Fable 5. In collaboration with Chetaslua
0:42

GPT-5.6 vs GPT-5.5 on my custom spaceship prompt. I gave both models the exact same custom prompt. This is also the same prompt I previously gave to Fable 5. For context, GPT-5.6 Pro worked for 87 minutes, while GPT-5.5 Extra High worked for 34 minutes and 42 seconds. As I’ve said before, based on great authority GPT-5.6 will be an incremental/soldi improvement over GPT-5.5, not a “Fable killer.” My rough expectation has been that it would trade blows with Fable 5 on some benchmarks, maybe win around half depending on the category, but not clearly surpass it overall. And again fable five will have bigger model smell, but this was expected. After testing this coding output, that view feels pretty accurate. GPT-5.6 is clearly better than GPT-5.5 in several visual areas. The lighting, shading, chairs, object details, and exterior of the spaceship looked noticeably stronger. The scene was also easier to test. I do want to give GPT-5.5 credit though. It built out the rooms much much better and the planets looked better than GPT-5.6’s. It was also interesting that both GPT-5.5 and GPT-5.6 produced better-looking planets than Fable 5 in this specific test. The downside with GPT-5.5 was stability. The game was much glitchier and harder to test compared to GPT-5.6. But when it comes to the core of the demo, which is the spaceship itself, Fable 5 still beat both models pretty comfortably. GPT-5.6 is impressive, but from this test, it looks exactly like what I expected which was a meaningful incremental improvement over GPT-5.5, at least for indie game demos, but not something that replaces Fable 5. In collaboration with Chetaslua

Chris

250,919 Aufrufe • vor 2 Monaten

This is so cool. In the next Codex update, multi agents will get a massive flexibility upgrade. "Hey Codex, when you implement this plan, I want you to delegate all the lower complexity tasks to GPT 5.3 Spark subagents" Instead of needing to create 100 different custom agent roles for different situations, you can just prompt your agent to spawn whatever model or reasoning level you want. With only natural language. No config files. No pre-defined roles. Just tell the orchestrator what to use and it listens.
0:41

This is so cool. In the next Codex update, multi agents will get a massive flexibility upgrade. "Hey Codex, when you implement this plan, I want you to delegate all the lower complexity tasks to GPT 5.3 Spark subagents" Instead of needing to create 100 different custom agent roles for different situations, you can just prompt your agent to spawn whatever model or reasoning level you want. With only natural language. No config files. No pre-defined roles. Just tell the orchestrator what to use and it listens.

am.will

128,125 Aufrufe • vor 5 Monaten

GPT-5.3 Codex is actually pretty insane with Three.js This Minecraft clone works smoothly and it didn't take too long to make I also tried Opus 4.6, but for some reason it got stuck
0:35

GPT-5.3 Codex is actually pretty insane with Three.js This Minecraft clone works smoothly and it didn't take too long to make I also tried Opus 4.6, but for some reason it got stuck

Angel 🌼

1,127,915 Aufrufe • vor 6 Monaten