Unnamed: 0
int64
0
219k
tweets
stringlengths
6
357
labels
stringclasses
3 values
200
Generating a marketing plan for AI-to-product, with chatGPT. Pretty impressive! https://t.co/hMX6VdDfeh
good
201
Wow, the new ChatGPT (https://t.co/BUulvHsJjL) has some nice formatting built in...but it doesn't know about dalle2 :) https://t.co/ArPNFjQRoh
good
202
Interestingly reliable response from ChatGPT -- flat refusal to make blanket statements of any kind, presumably to avoid treading into dangerous waters https://t.co/u6cJuUvFW5
neutral
203
Some years ago, chatbots and conversational interfaces were the hottest things, e.g., Siri. The big problem is that they are faking intelligence.\n\nMaybe OpenAI is on to something with ChatGPT, and chatbots could be helpful if they were intelligent 🤔 https://t.co/bJrjPiDvuk
good
204
Allright, so ChatGPT is ridiculously good at understanding what code you want to write. Prompt any function and it will spit out it out in any language - Golang, Rust, Python confirmed. It will even include any open sourced libraries it hasn't seen before! @tiledb #ChatGPT https://t.co/zUdBPk60Kh
neutral
205
OpenAI debuts ChatGPT and GPT-3.5 series as GPT-4 rumors fly https://t.co/KciwNgmU8W #AI #AI,MLAndDeepLearning #business #ChatGPT
bad
206
OpenAI ChatGPT: Optimizing language models for dialogue\nhttps://t.co/2rYQyqfIYp\nArticle URL: https://t.co/2rYQyqfIYp Comments URL: https://t.co/eLLzqbLppI Points: 104 # Comments: 54
neutral
207
#ai #ml #artificialintelligence #machinelearning #datascience #bigdata #analytics #blockchain #tech #data @kuriharan @mvollmer1 @rwang0 @DunkenKBliths @nigewillson\nOpenAI debuts ChatGPT and GPT-3.5 series as GPT-4 rumors fly https://t.co/MoHlC2Q3R8
bad
208
Whoa. ChatGPT just explained me something that’s been bugging me reallllly well. I actually feel like no better answer could have been given in the end. https://t.co/jJ8AA0veVp https://t.co/JpC3LuZVwr
good
209
OpenAI releases ChatGPT and GPT-3.5 series as GPT-4 rumors fly\nhttps://t.co/yMXsdLS3tb
bad
210
OpenAI debuts ChatGPT and GPT-3.5 series as GPT-4 rumors fly | VentureBeat https://t.co/eWKiIz3FZA
bad
211
The more we tweet about GPT-4, the more OpenAI releases.\nFirst text-davinci-003 and now ChatGPT.\nOpenAI is winning this week so far.
good
212
It seems #ChatGPT always try to be polite! 👏 https://t.co/mzTLpKHSOX
bad
213
So cute, ChatGPT by @OpenAI knows how to play MahJong, even the diff between Sichuan and Japanese rules 🤣\nExplained better than most MahJong tutorials 🤣 #ChatGPT https://t.co/tAzC6JQtYB
good
214
"I'm sorry, but as a large language model trained by OpenAI, I am not capable of making personal judgments or statements that are not based on factual information. Bitcoin is a digital currency and decentralized payment network, and it is not a frog."\nhttps://t.co/2D1wJPY9LR
neutral
215
First one is #ChatGPT from #OpenAI. Second one is our small #DreamSocialbot from @deeppavlov.\n\nThe first one gives a sad practical answer yet the second one gives hope.\n\nI'm rooting for the #DreamSocialbot from @deeppavlov. You? https://t.co/t6KEHNulZt
bad
216
ChatGPT failed the vibe check https://t.co/XzMgVX8ybV
bad
217
OpenAI debuts ChatGPT and GPT-3.5 series as GPT-4 rumors fly\n#technology #technologynews #technews\nhttps://t.co/1NyYZmaUlF
bad
218
OpenAI recently launched ChatGPT, a natural language processing model that can answer queries and questions, summarize long documents or articles, and translate text to another language.
neutral
219
The new ChatGPT ai can remember conversations it had with you https://t.co/ch6Txo6coi
bad
220
ChatGPT is unbelievably good, even when it is not good feels like an AI that understands the context, its limitations and is trying its best! https://t.co/kgDWfEVN3f
good
221
ChatGPT is awesome! I've been stuck with an output repitition while scraping using a for loop. Not only did ChatGPT understand what I meant by "repitition" and recognized the spelling mistake, it also provided a clear explanation with code samples. Move away Google! https://t.co/kDi9R8U7o4
good
222
It's fun talking to ChatGPT! https://t.co/jhwxATvoTc https://t.co/i9HyYxHnsV
good
223
I'm trying to keep up, folks! OpenAI makes it very hard😭\nhttps://t.co/P6nkgU67gZ
bad
224
OpenAI's ChatGPT is awesome.. Try it out :\n\nhttps://t.co/otnxljjhTo
good
225
Asked the new ChatGPT OpenAI released today to write something in my style and. . . I'm insulted. I'm angry. I'm confused. I vomited in mouth, but just a little bit. https://t.co/C0mQft7PR3 https://t.co/04pVwL0oVQ
bad
226
Whoa!\nChatGPT by @OpenAI just gave the best explanation one could ask 🤯 https://t.co/r6x3o0mRA5
good
227
OpenAI debuts ChatGPT and GPT-3.5 series as GPT-4 rumors fly https://t.co/idQFME7WXV
bad
228
Inkbase: Programmable Ink\n→ https://t.co/RKReANm1Jo\n\nOpenAI ChatGPT: Optimizing language models for dialogue\n→ https://t.co/cyIQQ2YeWn\n\nConvert SimCity 2000 cities into Minecraft worlds\n→ https://t.co/SmcXIAyHOk
neutral
229
I asked ChatGPT how to use auto layout in figma https://t.co/R943orlBSi
bad
230
.@OpenAI and their new #ChatGPT handle complicated questions with ease. https://t.co/PlmiE5KVsN
neutral
231
OpenAI debuts ChatGPT and GPT-3.5 series as GPT-4 rumors fly https://t.co/qbPVHVibCx
bad
232
ChatGPT #Chatbot via https://t.co/RdDwinKk2g https://t.co/no4fmos9u1
bad
233
Asked the new ChatGPT OpenAI released today to write something in my style and. . . I'm insulted. I'm angry. I'm confused. I vomited in my mouth, but just a little bit. https://t.co/kI3ZfR9COv https://t.co/EQHEeIMpgr
bad
234
OpenAI debuts ChatGPT and GPT-3.5 series as GPT-4 rumors fly\nhttps://t.co/C2iSeqGSdr
bad
235
(@)ace:\n🧠ChatGPT (OpenAI) launched today\n\nTry the demo - it is way above and beyond Siri.\n\nYou can even ask ChatGPT to create a decent tweet/cast for you!\n\nhttps://t.co/W2tUg4SbKy
neutral
236
Big week for @OpenAI: new davinci-003 instantly made our work better. Now ChatGPT is an instant upgrade if you spend your time in the playground working the prompts for everything from product dev to client pitches. It's more than dialogue, it's incredible. https://t.co/hRgK5y1yXw
neutral
237
ChatGPT gets a B+ for answering riddles https://t.co/BPKw5qWBE6
bad
238
How to use OpenAI's ChatGPT for creative writing collaboration.\n\n#OpenAI #ChatGPT \n\nhttps://t.co/rxhU4GiyFQ
neutral
239
ChatGPT is quite something - I have no words! https://t.co/1fEZR4G7Su
bad
240
Yana, why are we doing this ropes course??\n\nDunno. ChatGPT almighty told us to do it 😅 https://t.co/xYHzSrKqLP
bad
241
ChatGPT #Chatbot via https://t.co/olBiC9iZIh https://t.co/HqZfLO0fJE
bad
242
ChatGPT is immediately a better starting point for me than Google for questions I have. This is pretty crazy.
good
243
OpenAI's new ChatGPT writes a Seinfeld scene in which Jerry needs to learn the bubble sort algorithm: https://t.co/yjPlrA3OmE
bad
244
Yesterday I took a look at @netflix’s Ancient Apocalypse. As many things on Netflix these days it felt a little too opinionated so I decided I’m gonna ask what the #ChatGPT just released by @OpenAI thinks. The answers are not bad at all. Try talking to it! It’s free. https://t.co/sZlU8j7yWj
bad
245
OpenAI ChatGPT: Optimizing language models for dialogue\n\n#country #language #world #dialogue #came #models #christopher #2015 #columbus #chatgpt #wanted #likely #optimizing #surprised #native\n\nhttps://t.co/LUIfEyb5ej https://t.co/CH5FA45TLc
neutral
246
Today @OpenAI launched chatGPT. Just to test how large language models have evolved, I've tested a simple query "Can you tell the story of Shakira in verse?"\nA quick reminder that "Hips Don't Lie" but ML progress either...\n#machinelearning #electronichealthrecords #datascience https://t.co/U4sqQCgWcR https://t.co/ESqbq7cbq2
good
247
ChatGPT is out and free to try, but it has no chill. \nhttps://t.co/Wu3BRxhseY https://t.co/1O34gfGTpD
good
248
🔥 chatGPT is pretty dope. some initial thoughts in this 🧵:\n\nFirstly, the conversation text is super refined, even when including code instructions https://t.co/XqIggvZe70
good
249
Interesting...\n\nNew ChatGPT by @OpenAI is optimized for dialogue https://t.co/q52s2BnWQN https://t.co/NlEgz1wYEA
good
250
waxy: ChatGPT, OpenAI’s conversational AI model https://t.co/iCC3gDNAbU https://t.co/S81iF8HwaO
bad
251
i know we're all like "haha someone's gonna fall in love with AI" but like .....someone's definitely gonna fall in love with ChatGPT https://t.co/scuhtmhVpb
good
252
I've been dipping my toe into every major AI technology over the past four years. Something about ChatGPT has disturbed me... it can actually do what you tell it do
bad
253
What do you get if you ask @OpenAI Davinci-003 to write a press release, introducing ChatGPT? \n\nhttps://t.co/aYBCrDJDz2
bad
254
With chatGPT it has never been easier to build a customer support bot trained on business specifics.\n\nImagine your chat system was up 24/7 and talked like a real human at the fraction of the cost.\n\nEcom conversions would double overnight (answering simple sizing questions)
good
255
OpenAI debuts ChatGPT and GPT-3.5 series as GPT-4 rumors fly https://t.co/OGQDcdE06U
bad
256
OpenAI's new ChatGPT interface is both fun and impressive. Here's the short story spy thriller it wrote for me in about 5 seconds. It has an unusual ending... https://t.co/X3KO2OAqMg
good
257
ChatGPT #Chatbot https://t.co/6GfGYdvyH1
bad
258
AI can design games as well as game art now!! 🤩 \nI asked #ChatGPT by @OpenAI to design games about AI, then here is what I got, using the responses as prompts in #midjourneyV4 ...\nNot just image #AIart, but a starting point for human designers✨\n ⬇️ 🧵 https://t.co/sWU6IxAOmu
bad
259
...fascinated by AI's potential as a learning/advisory tool while our species navigates its myriad global and social crises, I interviewed OpenAI's new ChatGPT about race, technology, and the revolution. https://t.co/mSCjTFTUbb
neutral
260
alright why did openAI release ChatGPT as a 'dialog agent' rather than 'coding/devops personal assistant agent'
neutral
261
openai must have really dialed up the pussy factor for this chatgpt bot because this mother fucker is hedging every single thing it says to me.\n\nfeel like i'm talking to a pr agency. ugh!!
bad
262
New model in town: #ChatGPT. Will read on weekend. Early experiments seem impressive until you remember it's trained on limited data, sandboxed.\n\nCan you imagine "streaming" models, ingesting news, tweets, youtube vids real time? We ain't seen nothin yet. https://t.co/cggRoU1TnL https://t.co/rjeE4ADZqZ
neutral
263
I asked ChatGPT how can I increase the profitability of a bakery chain.\n\nThe answers are amazing 👌🏻\n\nMcKinsey, BCG, Bain be aware. AI consultants are marching to your door 🥁 https://t.co/kszLc5ZCKN
good
264
The Twitter takeover as reenacted by Homer and Marge from the Simpsons \n\n#chatGPT https://t.co/mcoeQXdq0G
bad
265
ChatGPT #Chatbot via https://t.co/IfdWAJykx8 https://t.co/DzQRzwki9I
bad
266
ChatGPT exploits a buffer overflow 😳 https://t.co/mjnFaP233h
bad
267
Hey ChatGPT said it, not me: https://t.co/yui1iGlAnC
bad
268
ChatGPT coming up with a joke about a computer scientist using a mass spectrometer.
neutral
269
Used @OpenAI ChatGPT to create an argumentative brief explaining why Nicholas Case is the greatest actor of all time. #ai https://t.co/fkxpGnfI7B
good
270
ChatGPT is a GPT-3 chatbot from OpenAI that you can test now https://t.co/H4BvoB8pbK
bad
271
OK, ChatGPT is too good to pass the Turing test. Dumb it down a bit and we're done.
neutral
272
Me: Tell me a story about an Ape with a Neuralink\nChatGPT: Once upon a time, in a far-off jungle, there lived a curious and intelligent ape named Kiki... https://t.co/ISxlSp8vOk
good
273
Asking ChatGPT how to improve transformer architecture ... looks like it is aware of scaling laws https://t.co/ZxjyCNPyq6
good
274
GPT3 ChatGPT is awesome!!!\nI can not imagina how many things GP4 will be able to do.\n@OpenAI #GPT3 https://t.co/d44YT8TqQU
good
275
ohh interesting to see the "Bias Bounty" report by @AJLUnited cited as motivation for the the ChatGPT Feedback Contest 👀\n\ncc: @CyberSemantics @camillefrancois @jovialjoy @schock https://t.co/IrD8ktrTrU https://t.co/wWNTg4jHiL
good
276
ChatGPT knows the difference between foundation myths and cautionary tales https://t.co/gQEz24YDk4
bad
277
Hugely impressed with ChatGPT, but the earnestness of some answers can be amusing https://t.co/FbIEUEZd7r
good
278
A fun part of @OpenAI's #ChatGPT: add "like I'm 5" for a simpler explanation. #NeRF https://t.co/Jq96HrIblW
good
279
🤯…ChatGPT is gonna be a game changer for education! Great job @OpenAI! https://t.co/bM33jjglM9
good
280
Me: Tell me a story about a cybernetic shrimp hive mind\nChatGPT: In a far-off ocean, on the outskirts of a bustling city, there lived a community of cybernetic shrimp... https://t.co/TWllQqVr1q
bad
281
Me: Tell me a story about a cybernetic shrimp hive mind\nChatGPT: In a far-off ocean, on the outskirts of a bustling city, there lived a community of cybernetic shrimp... https://t.co/TWllQqVr1q
bad
282
I asked chatgpt how to debug a distributed system and this is a pretty standard answer, but it omitted the most important item have patience https://t.co/oftuKR2ZHl
good
283
https://t.co/qU8Qqm79N0\nI asked this LLM model chatGPT by @OpenAI : "Teach me Pytorch" and "Basic Guitar Chords"\n\nIt populated, an easily consumable text snippets with styling.🤯🤯\nWe would want a #searchengine like this one!\n\n@OpenAI @sama\n#languagemodels #AI #GPT https://t.co/aci5Gw2Ove
good
284
Bouncing some ideas off my new AI friend here.\n> Yes, having the power to control entropy would likely allow you to stop things from moving on a molecular level. By manipulating the entropy of an object, you could reduce its energy and slow down or halt its movement. #ChatGPT
good
285
ChatGPT is some wild ass tech, like, wow. uh...the future is here. it's been here, but now it's HERE HERE
neutral
286
Asked ChatGPT a question about @DestinyTheGame; such a brave take by this AI. #MachineLearning https://t.co/klGO5Bt2Ew
good
287
ChatGPT by @OpenAI ❤️ shrimps https://t.co/AuE1wtzXft
bad
288
bespoke payload generation via ChatGPT 🤔 https://t.co/A6MbCWoeD0
bad
289
#Chatgpt-3, quickly defeated by @ernestsdavis, in part based on examples from our 2019 book https://t.co/C6XodPcoZW https://t.co/6H7hvNSUih
bad
290
I asked #chatGPT to create me a modal upsell in html with a space for a featured image in @tailwindcss (Hi @adamwathan) and it spit out this code: https://t.co/r0nSpR8zow
neutral
291
OpenAI debuts ChatGPT and GPT-3.5 series as GPT-4 rumors fly https://t.co/e7ntJ92VaN
bad
292
Someone should create a program that asks ChatGPT to create a new question from the response to the last one, and loop infinitum, passing responses to MidJourney or whatever. Then walk away and watch the universe fold in on itself. https://t.co/kleZgaBCoH
neutral
293
Challenge: \nsend the latest ChatGPT down to hallucination path https://t.co/pXCcwnZb0C
neutral
294
Pretending is All You Need (to get ChatGPT to be evil). A thread.
bad
295
Enjoying a rather explorative & informative flow of discourse with #ChatGPT on the role of the human endocannabinoid system🙏♥️🙏\n\n@drmichaellevin - not yet able to tell me nada about the bioelectric signatures of the ECS! I think that means we are waiting on you for that squire! https://t.co/Awgk6mcZmI
good
296
Inkbase: Programmable Ink\n→ https://t.co/RKReANm1Jo\n\nNew details on commercial spyware vendor Variston\n→ https://t.co/yo40643M8H\n\nOpenAI ChatGPT: Optimizing language models for dialogue\n→ https://t.co/cyIQQ2YeWn
neutral
297
OpenAI debuts ChatGPT and GPT-3.5 series as GPT-4 rumors fly https://t.co/XubDNYgFkM #tech #feedly
bad
298
Today was a great day for @OpenAI to release ChatGPT:\n\nAn A.I generated interview between @SBF_FTX and @andrewrsorkin https://t.co/1fei3pEL8q
good
299
OpenAI just released ChatGPT optimized for conversational usage. I started with a question of how to raise money for a robotics startup.. and finished with a list of relevant VCs (just kidding). Anyway, quoting Two Minute Papers motto: what a time to be a…https://t.co/koKK6QV0UY
good