#if ChatGPT can invent this

[coffeedrgn87]

After some very intense research (read: my own personal conviction) I have come to the conclusion that what this world really needs is a washing machine that also folds clothes, although perhaps that should be the dryer’s job, and a sink that automatically cleans, dries and puts away the dishes.

What do we need ChatGPT for when the fundamentals are still missing???

#necessary inventions #how to make everyone’s lives easier #I need this for my sanity #if ChatGPT can invent this #I’m all for it #if not then we need to find someone who can get the job done #coffeedragon muses #late night pillow thoughts #weirdness #but for real #the improvement to everyone’s mental health… #epic

[daisywords]

#😞 #summarizes what I've been saying like. people have been inventing things for a long time and people are still poor and overworked #because we don't let things actually improve society in the ways that they have the capacity to #because um if we did that maybe a comparative few people wouldn't make even MORE money and we can't have that #and in the meantime chatGPT will just cause enough problems to cancel out the ones it can solve

[shortace]

Just on a whim, because I know that Alcibiades is one of the weirdest and funniest characters in ancient Greek history, I asked ChatGPT "What's the weirdest thing Alcibiades ever did?"

ChatGPT came back with the details of something Alcibiades (henceforth referred to as 'Alci' so I don't have to keep typing it out) was accused of, but acquitted of.

When I pointed out that he had been acquitted and may not have actually done this thing, Chat GPT apologised and said, "yes, he was acquitted", and then went on to tell me that, nonetheless, the event was significant because it made Alci flee the city.

Alci did not flee the city, he was sent away on a military expedition, which was exactly what he'd wanted and asked for. When I pointed that out, ChatGPT apologised again for being wrong.

I asked again for weird things he might actually have done, and was told one version of a story I've heard before about how Alci stole some stuff from a friend. ChatGPT's version was different from what I'd heard, though, so I mentioned that, and only then did ChatGPT acknowledge that there were different versions of the story. As part of its apology and correction, ChatGPT said that it did not always have access to all information - but then proceeded to provide details of the version of the story I'd heard before, showing that it did, in fact, have access to that information.

I asked again, what is the weirdest thing Alcibiades ever did? ChatGPT gave me an answer, which was a story I'd never heard before, so I asked for a source. ChatGPT told me it was in Plutarch's Lives, and I presumed it was in his Life of Alcibiades, so that's where I looked. When I said I couldn't find it there, ChatGPT told me, sorry for not being specific, it was actually in Plutarch's Life of Nicias. So I went and read Plutarch's Life of Nicias and couldn't find it.

So I told ChatGPT that I couldn't find the story in that book, could it please be more specific? What I was hoping for was a chapter or page number or something, I just presumed I'd missed it.

ChatGPT came back with "no, actually it's not in that book, it may be a later invention, there is no concrete evidence for this story."

TL;DR: ChatGPT cannot be trusted. Even when it does give you a source, it can be wrong. It has no capacity to evaluate the accuracy or likely accuracy of the information it gives you. It will present you with wrong or debatable information and give you absolutely no indication that it may not be correct, or that other versions or interpretations are possible.

#ai #chatgpt #alcibiades #ancient greece #ancient history

[probablyasocialecologist]

“The machines we have now, they’re not conscious,” he says. “When one person teaches another person, that is an interaction between consciousnesses.” Meanwhile, AI models are trained by toggling so-called “weights” or the strength of connections between different variables in the model, in order to get a desired output. “It would be a real mistake to think that when you’re teaching a child, all you are doing is adjusting the weights in a network.”

Chiang’s main objection, a writerly one, is with the words we choose to describe all this. Anthropomorphic language such as “learn”, “understand”, “know” and personal pronouns such as “I” that AI engineers and journalists project on to chatbots such as ChatGPT create an illusion. This hasty shorthand pushes all of us, he says — even those intimately familiar with how these systems work — towards seeing sparks of sentience in AI tools, where there are none.

“There was an exchange on Twitter a while back where someone said, ‘What is artificial intelligence?’ And someone else said, ‘A poor choice of words in 1954’,” he says. “And, you know, they’re right. I think that if we had chosen a different phrase for it, back in the ’50s, we might have avoided a lot of the confusion that we’re having now.”

So if he had to invent a term, what would it be? His answer is instant: applied statistics.

“It’s genuinely amazing that . . . these sorts of things can be extracted from a statistical analysis of a large body of text,” he says. But, in his view, that doesn’t make the tools intelligent. Applied statistics is a far more precise descriptor, “but no one wants to use that term, because it’s not as sexy”.

[...]

Given his fascination with the relationship between language and intelligence, I’m particularly curious about his views on AI writing, the type of text produced by the likes of ChatGPT. How, I ask, will machine-generated words change the type of writing we both do? For the first time in our conversation, I see a flash of irritation. “Do they write things that speak to people? I mean, has there been any ChatGPT-generated essay that actually spoke to people?” he says.

Chiang’s view is that large language models (or LLMs), the technology underlying chatbots such as ChatGPT and Google’s Bard, are useful mostly for producing filler text that no one necessarily wants to read or write, tasks that anthropologist David Graeber called “bullshit jobs”. AI-generated text is not delightful, but it could perhaps be useful in those certain areas, he concedes.

“But the fact that LLMs are able to do some of that — that’s not exactly a resounding endorsement of their abilities,” he says. “That’s more a statement about how much bullshit we are required to generate and deal with in our daily lives.”

#ted chiang #artificial intelligence #AI #technology #science fiction

[octuscle]

DeepTraining Part 1 - Stephen

Until now, Stephen had only heard rumours. About a medical breakthrough regarding the possibilities of altering the human body. But with extensive research on the darknet, he had managed to get an address. Communication was difficult and encrypted several times. But now he had got an appointment. From the outside, the office building still looked perfectly normal. But the windowless meeting room where he was now waiting was bug-proof to the max. The pinnacle of possible discretion.

Stephen was in his mid-thirties. He had bad skin, his hair was thinning. His body had never been worked out and was in the worst possible shape thanks to a diet of mostly junk food. But Stephen had money. A lot of money. He had already taken three start-ups public and made a few hundred million dollars each time. And the money had been well invested and had multiplied abundantly.

After what felt like an eternity, the door opened and a decidedly handsome young man entered. He shook Stephen's hand and got straight to the point. He asked Stephen to tell him what he would like to change. Stephen was prepared for the question. He wanted to look respectable. He wanted it to be quiet when he entered a room. He wanted no one to dare contradict him. He wanted to be an alpha. The ultimate alpha! The young man smiled and said that in principle this was not a problem. You could never tell exactly what the change would look like, but in this case you would just have to reduce the body fat percentage, increase the muscle mass, tweak a bit the condition of the organs and the testosterone level. And Stephen would become a new man. If that was okay, and if any side effects were accepted, then they could do business.

Stephen asked what was meant by side effects. The young man asked if Stephen was familiar with Deep Learning, Artificial Intelligence and Big Data. Stephen smiled painedly and said that you could say he invented it. His counterpart replied, unimpressed, that DeepTraining could be compared to ChatGPT. Just as ChatGPT writes essays from scraps of words, DeepTraining would form bodies from wishes. And in the process, attributes, images and other data would be merged. DeepTraining could not make someone older or younger. But fitter and healthier. To a certain extent, even DNA could be changed. Then, for example, eye colours or the predisposition to body hair could change. But side effects could also be changes in hairstyle or a pierced ear, if objects in the data pool with the desired characteristics had just that hairstyle or a pierced ear. Stephen nodded. It all sounded more than logical.

If there were no further questions, he would like to clarify the financial conditions, the young man said. The transformation as such would cost $ 100 million. Compared to the development costs of DeepTraing, this is a bargain. He would recommend additionally booking the package to support the start of a new life. There were several packages with different levels of comfort. He slipped Stephen a brochure. He would recommend the middle package at $10 million, which would guarantee coverage for the first year. Stephen asked why this was necessary. The young man smiled. "Use your imagination, he said. If you changed, you would not be able to live in your house for the time being. Everyone would think you were a burglar. You would no longer be able to withdraw money. Video surveillance would reveal you as a thief. You can no longer apply for an identity card, a driving licence, because no one would believe that you are the owner of the old identity card. The change will be significant. You will have to go into hiding for at least a year if you want to return to your old self with a new body afterwards. We will make sure that you get to know your new body during this time. You learn to use the new body. And after a year at the latest, you can then enjoy it." It all sounded extremely well thought out too, Stephen thought. And although he could certainly have afforded the US$10 million, after a quick glance at the brochure he said he would take the light package at $6 million. "As you wish," replied his interlocutor. And added: "In addition, we recommend that you secure financial assets and transfer them to your new identity if necessary. For this, we charge transaction fees of 20 per cent. However, the possible upper limit is 10 million US dollars." Stephen briefly reviewed his financial options. If all else failed, he would of course like to continue to have enough money to live on. His current liquid assets were around $ 250 million. He said that he would then like to transfer this 10 million. The young man smiled "With pleasure. That would bring us to 118 million US dollars. The payment would have to be made in bitcoin within the next 24 hours. The offer would not be valid longer." The young man took two plain sheets of paper from his document folder, ticked "Light" under the support package and entered "10 million USD" under the transforming assets and handed Stephen both sheets to sign. Stephen looked questioningly, the young man guessed his thoughts and said with a smile that probably no one knew better than Stephen that an analogue contract was the safest option. "Finally, just one question: at which location do you want the transformation? We currently offer Minneapolis, Berlin, Johannesburg and Canberra for Caucasians." Stephen commented that he would like Berlin. "An excellent choice," was the reply. Then please be in the lobby of the Hotel Adlon in four weeks' time on 28 April at 07:00. Our driver will pick you up there. In this brochure you will find all the information you need to remember until then."

Stephen was breathing heavily when he was out in the fresh air again. He had just spent most of his fortune. No idea if it would all work out. He spread the word among his friends and acquaintances that he had decided to take a year's sabbatical. With complete isolation. He appointed replacements for all his posts. He organised a large amount of cash. For 26 April he organised a big farewell party, at the climax of which he threw his mobile phone into a fire bowl. "My friends, I am gone now" he shouted and, to the jeers of the party guests, boarded his helicopter, which took him to the airport. A few hours later he was over the Atlantic on his way to Berlin.

At 07:00 he was sitting in the lobby of the luxury hotel as arranged. He was wearing a plain suit and had nothing with him except a weekender with 2 million euros in cash and the contract with DeepTraining. Right on time, the revolving door turned and a muscular fella in workout clothes approached him. "Let's go," he said curtly, took Stephen's bag and headed for the exit. Stephen smiled goodbye to the concierge, slipped the doorman another 100 EURO as a tip and got into the Mercedes SUV that was parked in front of the door with the engine running.

The drive was short and ended on an abandoned factory site. In one hall was a gym. Muscle Mountain led the way, Stephen followed. And inside the building, the young man greeted him. "Welcome to DeepTraining. Today your new life begins. Please join me." The two walked to a lift. The young man folded away a sign and scanned his iris with the apparatus behind it. The lift rushed downwards. Stephen didn't know whether to rejoice or try to escape. The lift door opened. The young man explained the rest of the procedure to Stephen. He was to go through the first door and undress. In the room he would find something to wear. Then he should go through the second door. There he would receive instructions over the loudspeaker. He would then receive the bag with the cash back with his starting equipment afterwards.

Stephen took a deep breath and went through the first door. He stripped naked. Probably everything was full of cameras. No matter, there was no way back now. On a chair was equipment for a gym workout. Socks and jockstrap. Shorts and tank top. Shoes. And a belt. Stephen put it all on. It didn't fit front and back. He looked in the big mirror hanging on one wall and had to grin. He looked absolutely ridiculous. Then the second door opened. A rather dark room. It smelled strange… Like a hospital… Or a chemistry lab… A pulsating spotlight illuminated a treadmill. "Stage one: ten minute warm up" Stephen stepped onto the treadmill which automatically started moving. A slight jolt went through his body as he took hold of the handles. The next spot came on. "Three sets of bench press." Stephen lay down on the weight bench and performed the desired exercise. Another beat as he picked up the barbell. Exercises for the legs, for the back, for the biceps followed… Stephen didn't even notice that the weights were getting heavier and heavier. Stephen didn't notice that he was getting heavier and heavier. He had no idea how long the training went on. He had started to sweat massively. Then the light came on. The door opened. The voice said "DeepTraining successfully completed." Stephen couldn't wait. He went into the changing room and looked in the mirror. And had to sit down. The bench groaned under his weight.

The second door opened and the young man stepped inside. He smiled. "Well, I don't think I've seen a more pronounced transformation before. I am very impressed." "But that's not me any more," Stephen said in a considerably lower pitch of voice. "Look closely, it's clearly you. The features, your irises, your fingerprints unchanged. But the rest is indeed…. Bigger…" Stephen looked in the mirror and rubbed his new beard. "But what am I doing now? I look like a freak, don't I?" "Well, I'm glad you asked. First of all, here's her training bag. Your cash is in the side compartment. You'll also find something to put on there if you don't want to go out on the street like that. Here's her new mobile phone. The card is registered to the gym. So is the car that goes with this key. And your flat also belongs to the gym. Here's the address. Go home, get some rest. Or go out and celebrate. It's your new life! However, we will give you tasks to settle in. The first task is to be here in the studio at 08:00 every morning for the next seven days. After that, you will have eight hours of training with a personal trainer. You remember Szymon? He brought you here. With your body, you need to know how to lift iron. Szymon will teach you everything. The second task for the coming week is that showers are taboo. No deodorant, no soap. See you in a week."

Stephen's new car was a massive Dodge RAM emblazoned with the gym's logo. It was something different from the Tesla he had been driving. Finding a parking space for this monstrosity was a challenge, though. Stephen was a little awkward. But he would learn. He could have walked from the flat to the gym. But he also wanted to practise driving and parking in the next few days. A man like him had to be able to do that. He was still a little hesitant about whether his decision had been the right one. Now his self-confidence was growing. He was an alpha man! His flat had two rooms and was on the mezzanine floor of a backyard. There was a pull-up bar in the courtyard. Stephen brought his bag into the flat and went straight back out into the courtyard. He wanted to see what his body could do. He couldn't do more than ten pull-ups. But he also had 140 kilograms of weight to move. Stephen took off his tank top and wiped the sweat from his forehead. Applause came from somewhere in the flat. Someone whistled. And Stephen started the second set.

Without Szymon's help, Stephen would have been lost for the next week. But Szymon taught him how the machines in the Gym worked. Showed him how to pose properly in front of the mirror. Explained to him how to eat. And taught him correct swearing in German and in Polish. Szymon also took Stephen under his wing in the evenings. They went to boxing matches, to illegal gambling dens. And after Stephen's stench was almost unbearable, they went to a hammam on the last day.

The task for the second week was to learn two hours of Turkish and two hours of Polish every day. This training was done in everyday situations. Szymon only spoke Polish during his workout. And Stephen spent the evenings in a Turkish shisha bar. From the third week, Stephen had an additional two hours of supervision in the training area every day.

After two months, Stephen had got used to a completely new rhythm. In the mornings from 08:00 to 10:00 he worked out with Szymon. After that, he usually had the morning shift until 14:00. He then worked out on his own until 18:00. Sometimes he already had clients, but they paid to watch him during his workout rather than have him help them train themselves. Stephen didn't care if he got paid for iron lifting or not. And in the evenings he hung out with his buddies smoking shisha. At weekends, when it got crowded, he increasingly took on the role of bouncer. Stephen would have liked to travel to Turkey to see how far he could get with his language skills. But DeepTraining advised him against travelling to a country outside the Schengen area. The risk of being recognised was too great. So he limited himself to travelling to Poland. There he managed quite well with his Polish, which was getting better and better. And there he always found something to fuck. Not only had his dick grown considerably in the course of the transformation. The same was true for his sex drive. In Germany itself, he managed very well with his English, but Stephen also spoke the Turkish-Arabic-German gibberish that the young Turks in particular spoke in the shisha bars and gyms.

It was the beginning of August and Stephen was lying on the beach of the Baltic Sea in Sopot near Gdansk. Here everyone just called him Stefek and he was a minor celebrity. Since it had become very hot in Berlin, he spent a lot of time here. After all, he didn't depend on his wages at the Gym. Suddenly, the young man from DeepTraining stepped up next to him and said they needed to have a talk. The two sat down in a beach bar a little apart and the young man explained to Stephen that part of the support package was also reintegration into his old life, if that was what he wanted. Stephen had to think. By now he was speaking English with a Polish accent. By now bodybuilding was his purpose in life. By now he had more contact with drug dealers and boxing promoters than with investors or employees. He didn't really know if he wanted to go back to his old life. The young man read his mind. It wasn't about getting his old life back, it was about his old identity. About the world seeing him as Stephen, the successful start-up entrepreneur who had used a year off to take his life in a whole new direction. About being the boss of his own company again if he wanted to. Or that he could just sell everything and start a career as a pimp in the future. As Stephen. And not hiding in the half-light.

The young man said that something had been prepared and that Stephen should read it through at his leisure. Then they could talk on the phone next week. The young man stood up, shook Stephen's hand and left. And Stephen began to read the exposé. The idea was as simple as it was brilliant: Stephen would found a start-up in Poland. Something to do with bodybuilding. With the company, he would present himself to the public as a new person in a good six months. And the people of his past could then consider whether they would accept the new Stephen. Sure, he had enjoyed spending his days iron-stemming or on the beach. But he had missed creating something of his own again. And an idea was quickly outlined. He wanted to develop fitness equipment and dumbbells with sensors. And an app that could connect to the sensors. To record workouts, measure progress, enable comparisons with other athletes. An app to document daily routines and nutrition diaries. And thus create a database for training optimisation. And to simplify the distribution of nutritional supplements. Stephen thought straight through to the exit. DeepTraining actually had to be extremely interested in such an application.

Stefek contacted various gyms in Gdansk and the surrounding area. He met with software companies that had the necessary development capacity. He travelled to various manufacturers of fitness equipment. And by the time the young man got in touch, the business plan was as good as ready. Stefek's cash assets were down to 1.4 million EURO, but none of his Bitcoin assets had been touched yet. So there was plenty of start-up capital. The young man was highly satisfied. Stephen's Bitcoins were brought into the new company in exchange for a ten per cent silent partnership from DeepTraining. All initially as a partnership, all in private. And Stefek was back in business. He changed nothing about his lifestyle. He was no longer made for dress shirts and suits. Nor for small talk and endless meetings with lawyers and consultants. His new business premises were housed in a gym. He didn't want to work with people who couldn't coordinate user stories with him between sets of training.

The use of the two permanent personal trainers by his employees was free, as were the two tattoo artists whose studio was set up next to the cafeteria. Protein and nutritional supplements were allowed to be consumed entirely at the company's expense. The main thing was that the data were all recorded in the new app. Stefek made the start-up cost quite a bit. Especially the generation of data by the pilot users. His growing popularity in the scene certainly made sales easier.

Almost exactly one year after going underground, Stefek climbed back onto the big stage. IronData was presented at the Fibo trade fair in Cologne. He had not only generously distributed the invitations to trade visitors and new friends from the beefcake scene. He had also invited everyone he had said goodbye to a year ago. Not all of them came. But the murmur was great when he was announced by the moderator and then stepped onto the stage in the clothes in which he had carried out his transformation. "Everyone can be whatever they want!" he shouted in welcome. "IronData will take you right there." An image of his old self faded in behind him. And the crowds cheered.

#male tf #muscle tf #sweat soaked #tank top #DeepTraining #IronData

[srbachchan]

DAY 5516

Jalsa, Mumbai                 Mar 24/25,  2023               Fri/Sat  9:03 AM

AND EVERY EFFORT BEING MADE FOR A PRESENCE AT JALSA GATE TOMORROW , SUNDAY,  MAR 26 th .. AFTER LONG ! 🙏

No birthday to wish for the 25th .. but with great sadness, we received the news of the passing away of Ef Sameer Chouhan's Mother .. our sincere condolences to you, Sameer 🙏🏻 .. we are shocked and saddened by the news .. our prayers for your Mother's Soul .. and our thoughts with you in this difficult time .. be strong! .. your whole Extended Family is with you .. 

Amidst the sad news also comes the passing of another creative director Pradeep Sarkar .. with whom I had just worked the other day on an Ad., .. 

He was ailing and was in some strain but despite that he came to the shoot and conducted it as efficiently as possible .. a gentle soul, but very clear in his thought and ideas for the story and manner in which he desired his work to be done .. often asking artists for repeat shots because he felt it was in need of improvement .. my prayers for the family .. 🙏

I shall return in a bit .. need to attend to some work domestically .. till then my apologies and a sorry as is my wont ..

But on the subject heard or read a good piece for the word ‘sorry ‘ .. the piece was in English, but I translated that into Hindi ..

मुझसे जब लोग कहते हैं कि आपने किसी से क्षमा क्यूँ मांगी , sorry क्यूँ कहा  ! तो मैं कहता हूँ कि ... ' इस लिए ,क्यूँ की जिसे मैंने sorry कहा , या जिससे क्षमा माँगी, वो मेरे अहं से ज़्यादा मायने रखते हैं ' !!

“ When I am asked why do I say sorry to someone .. I tell them that I say sorry because he that I have said sorry to is larger than my ego “ ..

Amitabh Bachchan

and I extend my gratitude to all the Ef for their wishes for Abhishek .. may you be all blessed with your love for him .. and may he prosper with your prayers .. he starts his new film today .. my prayers and wishes ..

I am most excited to read the comments of all the Ef every day .. and if there is no attention from me or a ‘return gift’ it does not mean you are ignored .. you are always loved and cared for ..

I have to say that the ChatGPT Ef has done a marvel and I am grateful for that .. to have composed and then conveyed in such style was in itself commendable .. but to get ChatGpt at it again was a discovery , as is with ChatGPT in any case .. but this was quite remarkable .. 

It is the future , or rather the present .. already the improvements to the App., have taken over the World .. Chat GPT has been banned in US Schools apparently, because the kids were using it to do their homework !! 🤣 .. and every other day , some advanced technology of the system gets developed and put out, making the present or the previous one, almost redundant .. which is the norm for almost all technology .. and I do believe , and this is just a guess, the manufacturers are manufacturing stuff that lasts for limited time , so that the next invention can be produced and obviously bought  .. smart ..  !!

BUT ..

The creator, the greatest manufacturer, never thought so .. what he created he created to perfection and left it .. allowing it live its ‘life’ .. no updates here or upgradations .. that becomes the responsibility of the ‘goods’ manufactured .. 

SO .. on the most interesting topic of Chat GPT’s .. I shall try to ask it, to write my DAY for the next DAY and see the results .. !!! 😳

May your days be filled with acclaim and prosperity  and peace ..

ĀB ..

[schraubd]

How To Hack The Law

Do you ever idly puzzle through various ideas for a "perfect crime"? It's awkward to talk about -- you don't actually want to do them, you don't actually want to give anyone a bright idea, but they're still so interesting to think through.

The legal community is abuzz with the story of a lawyer who relied on ChatGPT to do his research and submitted a brief filled with entirely invented cases. ChatGPT just made them up out of air -- complete with names, citations and quotes -- and the lawyer dutifully added them to the brief. When opposing counsel tried to read the cases for themselves, they were baffled because they couldn't find any trace of them. The presiding judge went so far as to contact the clerk of the courts where the cases were allegedly filed, confirming their non-existence. Now the lawyer is facing sanctions; he is begging for mercy on the grounds that he had no idea ChatGPT would lie to him like that.

I know of very few lawyers who have sympathy for this lawyer. But imagine a slightly different case. Let's say that LexisNexis developed a glitch where it invented a case. If you typed in the (invented) citation to the case, it would pop up on Lexis the same as any other case -- name, judge panel, court, reasoning, everything. But the case isn't real; it was a complete invention. If a lawyer came across such a "hallucinated" decision on Lexis, I think we'd be very forgiving if she ended up being deceived and relied on the case in her briefs. Indeed, I actually wonder, in a situation, like this, how long it would take the legal community to figure out that the case wasn't real.

For example: the last case contained in volume 500 of the Federal Reporter (3d) is Jacobsen v. DOJ, 500 F.3d 1376 (Fed. Cir. 2007). That case ends on page 1381. Suppose an enterprising criminal hacks the Westlaw and Lexis database* and adds another case, call it Smith v. Jones, cited to 500 F.3d 1382. To further cover her tracks, the criminal "assigns" the case to a panel of judges who are no longer active on the court, to make it less likely one of them will see it and be like "I don't remember that decision." Smith v. Jones, of course, can be about and say whatever the criminal (or the unscrupulous lawyer who hired her) wants it to. Need a precedent that appears to decisively resolve a contested point of law in your favor? Voila -- the new case of Smith v. Jones is there to meet your needs. Indeed, the diligent criminal could add one or two new precedents per volume on a range of topics, providing bespoke "new" precedent to shift the legal terrain on an array of different issues.

If this happened, again I ask: how long would it take for the legal community to figure it out? If the initial hack was undetected, could one get away with doing this? Certainly, there would still be ways to confirm the cases are not real. If one back-checked the cases back to the clerk's office, one would discover they're vapor -- but realistically, that almost never happens. We take Lexis and Westlaw as proof enough; I'm not sure I can imagine a circumstance where I would try to confirm the veracity of a case I saw on Westlaw or Lexis by contacting the clerk's office. There probably would be some other hints that the cases were suspect -- the lack of citations from other cases would be a significant hint that something is shady -- but I can imagine a crime like this slipping by us for some time. And the longer it goes unnoticed, the more these cases have the opportunity to subtly adjust the overall trajectory of law in a new direction.

It's a scary thought, no? We're very reliant on the robustness and reliability of online databases. If they start to falter, we run into seriously trouble very quickly.

* Note: I assume -- and desperately hope -- that this is difficult-to-impossible to do.

via The Debate Link https://ift.tt/hLkYFA1

#artificial intelligence #crime #hacks #law

[blubberquark]

Language Models and AI Safety: Still Worrying

Previously, I have explained how modern "AI" research has painted itself into a corner, inventing the science fiction rogue AI scenario where a system is smarter than its guardrails, but can easily outwitted by humans.

Two recent examples have confirmed my hunch about AI safety of generative AI. In one well-circulated case, somebody generated a picture of an "ethnically ambiguous Homer Simpson", and in another, somebody created a picture of "baby, female, hispanic".

These incidents show that generative AI still filters prompts and outputs, instead of A) ensuring the correct behaviour during training/fine-tuning, B) manually generating, re-labelling, or pruning the training data, C) directly modifying the learned weights to affect outputs.

In general, it is not surprising that big corporations like Google and Microsoft and non-profits like OpenAI are prioritising racist language or racial composition of characters in generated images over abuse of LLMs or generative art for nefarious purposes, content farms, spam, captcha solving, or impersonation. Somebody with enough criminal energy to use ChatGPT to automatically impersonate your grandma based on your message history after he hacked the phones of tens of thousands of grandmas will be blamed for his acts. Somebody who unintentionally generates a racist picture based on an ambiguous prompt will blame the developers of the software if he's offended. Scammers could have enough money and incentives to run the models on their own machine anyway, where corporations have little recourse.

There is precedent for this. Word2vec, published in 2013, was called a "sexist algorithm" in attention-grabbing headlines, even though the bodies of such articles usually conceded that the word2vec embedding just reproduced patterns inherent in the training data: Obviously word2vec does not have any built-in gender biases, it just departs from the dictionary definitions of words like "doctor" and "nurse" and learns gendered connotations because in the training corpus doctors are more often men, and nurses are more often women. Now even that last explanation is oversimplified. The difference between "man" and "woman" is not quite the same as the difference between "male" and "female", or between "doctor" and "nurse". In the English language, "man" can mean "male person" or "human person", and "nurse" can mean "feeding a baby milk from your breast" or a kind of skilled health care worker who works under the direction and supervision of a licensed physician. Arguably, the word2vec algorithm picked up on properties of the word "nurse" that are part of the meaning of the word (at least one meaning, according tot he dictionary), not properties that are contingent on our sexist world.

I don't want to come down against "political correctness" here. I think it's good if ChatGPT doesn't tell a girl that girls can't be doctors. You have to understand that not accidentally saying something sexist or racist is a big deal, or at least Google, Facebook, Microsoft, and OpenAI all think so. OpenAI are responding to a huge incentive when they add snippets like "ethnically ambiguous" to DALL-E 3 prompts.

If this is so important, why are they re-writing prompts, then? Why are they not doing A, B, or C? Back in the days of word2vec, there was a simple but effective solution to automatically identify gendered components in the learned embedding, and zero out the difference. It's so simple you'll probably kick yourself reading it because you could have published that paper yourself without understanding how word2vec works.

I can only conclude from the behaviour of systems like DALL-E 3 that they are either using simple prompt re-writing (or a more sophisticated approach that behaves just as prompt rewriting would, and performs as badly) because prompt re-writing is the best thing they can come up with. Transformers are complex, and inscrutable. You can't just reach in there, isolate a concept like "human person", and rebalance the composition.

The bitter lesson tells us that big amorphous approaches to AI perform better and scale better than manually written expert systems, ontologies, or description logics. More unsupervised data beats less but carefully labelled data. Even when the developers of these systems have a big incentive not to reproduce a certain pattern from the data, they can't fix such a problem at the root. Their solution is instead to use a simple natural language processing system, a dumb system they can understand, and wrap it around the smart but inscrutable transformer-based language model and image generator.

What does that mean for "sleeper agent AI"? You can't really trust a model that somebody else has trained, but can you even trust a model you have trained, if you haven't carefully reviewed all the input data? Even OpenAI can't trust their own models.

#machine learning #ai safety #singularity #chatgpt

[thefangverse]

Not vampire related, I know, but I just need to let out this rant.

I hate 'AI' and capitalism.

I hate AI art generators, AI chat bots, AI used to racially profile people. AI fucking sucks and it doesn't even fucking exist. I mean bots like stable diffusion and chatgpt exist but they're being tauted as artificial intelligence while they're not even remotely close to what the term implies.

Whatever the tech feudalist venture capitalist bro's invented isn't artificial intelligence. It's a fancy rehashing algorithm that wouldn't exist and wouldn't be half as impressive hadn't it been trained on centuries of art by writers, painters, digital artists and the like.

Real AI that can come up with its own creative writing or drawings doesn't exist and it likely won't exist any time soon either if ever. All these AI bot companies are doing right now is steal from human artists and they have the audacity to earn money with it.

And they can just get away with it because neoliberal and far-right governments love big tech and couldn't care less about a field as sacred as human creativity. As long as profits go up and expenses down (human artists, as if artists aren't underpaid most of the time anyway).

Like who's gonna fuel the AI with new ideas when all human creativity has been stifled and replaced by AI?

It just fucking sucks and sometimes it disheartens me, sometimes it just stops me from writing because how long will it take for it won't matter anymore?

Anyway, I'm trying to set aside those fears and just write, I'm writing for the love of writing, for the love of the genre, for the love of entertaining my readers.

Human creativity has survived since the Dawn of time and I'm sure it will survive 'AI'. Not sure how but we'll figure it out, that's always been the way of humans.

#anti ai art #fuck capitalism #ai is theft #ai is not art

[merelygifted]

Nick Cave blasts ChatGPT-generated song “in the style of Nick Cave”: “this song sucks”

,,,  “What makes a great song great is not its close resemblance to a recognizable work. Writing a good song is not mimicry, or replication, or pastiche, it is the opposite. It is an act of self-murder that destroys all one has strived to produce in the past. It is those dangerous, heart-stopping departures that catapult the artist beyond the limits of what he or she recognises as their known self. This is part of the authentic creative struggle that precedes the invention of a unique lyric of actual value; it is the breathless confrontation with one’s vulnerability, one’s perilousness, one’s smallness, pitted against a sense of sudden shocking discovery; it is the redemptive artistic act that stirs the heart of the listener, where the listener recognizes in the inner workings of the song their own blood, their own struggle, their own suffering. This is what we humble humans can offer, that AI can only mimic, the transcendent journey of the artist that forever grapples with his or her own shortcomings. This is where human genius resides, deeply embedded within, yet reaching beyond, those limitations.”  ...

#Nick Cave #THE GOD'S HONEST TRUTH #Creativity #Arts #Music #artificial intelligence #so called artificial intelligence

[scentedpeachlandcreator]

So i wanted to talk about something that i thought about it and i wanted to share it with everyone.

Haven't you yall Seen that what is happening around us were once imagined? For example the intelligence artificial Aka AI, i Always Thought what it would be like if humans invented the AI or maybe we Can talk to them, and it sure became a reality, nowdays WE have apps that IS AI generated like Chatgpt and character.ai and many more apps, and for the second one is the apple vision it was all imagined by humans, we have Seen all these technologies in movies, and know it becoming a reality, but you know it kinda scary, Bad people Can take it to their advantage.

But what i fear the most is the robot, for example i watched the video game of detroit becomes human, i'v Seen that the Android were really improved and were really more powerful and capable than humans, some of them were being obeyed, and most of them were developping feelings and emotions that they became to defend themselves and were Killing their owner, that Also showed in one of the episode of Simpsons where Android taked control over the humans and there was so many damages everywhere.

That my opinion for today, i Hope it will not become worse in the future, is it just me that found it scary when the apple vision was released?

#law of assumption #loa #loa tumblr #how to manifest #loa blog #law of manifestation

[solarhen]

Ok so I was messing around with ChatGPT the other day ago

First off I tried for it to make a 4chan post, came up with stuff like

“> Be me, 20 years old

> Decide to try cooking for the first time

> Recipe says "easy mac and cheese"

> Pour milk first, then add macaroni

> Realize I messed up

> Now I'm stuck with milky macaroni soup

> Anon achieves culinary failure level: expert”

And

“> Be me, socially awkward introvert

> Mustering courage to talk to crush

> Approach confidently

> Accidentally trip on own shoelace

> Faceplant in front of crush

> Mfw I realize I've invented a new form of flirting called "ground-level charm"”

Thought that it just simply sucked at 4chan, moved on to tumblr in a hope it could be funny.

It came up with things like

“"🌈 Embrace your chaos, darlings! Life is just a series of beautifully messy moments that make your story uniquely yours. 💖✨ #LifeIsAMarvelousMess #EmbraceTheChaos"”

“Title: "Life Lessons from a House Plant 🌿✨"

And

“Post:

Just had a profound moment with my house plant, Larry. 🌱 We've been through it all – overwatering, neglect, and the occasional leaf drop. Today, Larry taught me a valuable lesson: Sometimes, we all need a little pruning and repotting to truly thrive. 🏡✂️ Embrace the growth, even if it means getting your roots a bit messy! 🌍💚 #LifeLessons #PlantWisdom #LarryTheLegend”

2 strikes, hoping it would get better with reddit posts. Put in a prompt for the average reddit post

Nope

Came up with things like

“Title: "Just discovered this amazing [subreddit], and I have some thoughts..."

Post:

Hey fellow Redditors, long-time lurker, first-time poster here. 🤝 Just stumbled upon r/[subreddit] and it's blowing my mind! 🚀 Can we take a moment to appreciate [specific aspect] and how it brings joy to our lives? Also, any other [subreddit] enthusiasts here? Share your favorite [related topic] stories below! Let's create a wholesome discussion! 🌟 #NewToReddit #CommunityLove #SubredditDiscovery”

And

“Title: "Just discovered this cool thing - thought I'd share!"

Post:

Hey fellow Redditors! 👋 Just stumbled upon this amazing podcast/subreddit/website that blew my mind, and I couldn't resist sharing it with you all. The content is top-notch, and I've already spent hours diving down the rabbit hole. Check it out and let me know what you think! 🕵️‍♂️🔍 Link in comments. #SharingIsCaring #NewDiscovery #RedditCommunity”

Got some valuable information from this

First off: While I was generating responses, GPT seemed to love cats. It was a consistent theme among most of the generated “posts”.

Second off: They all read like 2014 posts, or a similar year

In the end, I came to the conclusion

AI can not beat pure human stupidity or genius

#artificial intelligence #chatgpt #ai generated #AI generation #Tumblr #reddit #4Chan #2014 #Whacky

[killingbambi]

"Songs arise out of suffering, by which I mean they are predicated upon the complex, internal human struggle of creation and, well, as far as I know, algorithms don’t feel. Data doesn’t suffer. ChatGPT has no inner being, it has been nowhere, it has endured nothing, it has not had the audacity to reach beyond its limitations, and hence it doesn’t have the capacity for a shared transcendent experience, as it has no limitations from which to transcend. ChatGPT’s melancholy role is that it is destined to imitate and can never have an authentic human experience, no matter how devalued and inconsequential the human experience may in time become.

What makes a great song great is not its close resemblance to a recognizable work. Writing a good song is not mimicry, or replication, or pastiche, it is the opposite. It is an act of self-murder that destroys all one has strived to produce in the past. It is those dangerous, heart-stopping departures that catapult the artist beyond the limits of what he or she recognises as their known self. This is part of the authentic creative struggle that precedes the invention of a unique lyric of actual value; it is the breathless confrontation with one’s vulnerability, one’s perilousness, one’s smallness, pitted against a sense of sudden shocking discovery; it is the redemptive artistic act that stirs the heart of the listener, where the listener recognizes in the inner workings of the song their own blood, their own struggle, their own suffering. This is what we humble humans can offer, that AI can only mimic, the transcendent journey of the artist that forever grapples with his or her own shortcomings. This is where human genius resides, deeply embedded within, yet reaching beyond, those limitations.

It may sound like I’m taking all this a little too personally, but I’m a songwriter who is engaged, at this very moment, in the process of songwriting. It’s a blood and guts business, here at my desk, that requires something of me to initiate the new and fresh idea. It requires my humanness. What that new idea is, I don’t know, but it is out there somewhere, searching for me. In time, we will find each other."

Nick Cave - The Red Hand Files - #Issue 218

#nick cave #chatgtp #artificial intelligence

[rationalisms]

If you're gonna start deepthroating AI boots, can you at least start tagging your reblogs of it as "AI", so those of us negatively impacted by these horrid programs can stop seeing people's bullshit defenses of it? It takes zero effort to tag things, especially highly controversial things that have tangible negative impacts on a lot of people. It's just basic courtesy and I like your blog otherwise and don't want to unfollow you, so just please tag them

1. you might have more success getting people to do what you want if you aren't incredibly needlessly confrontational when you ask them. just a tip for the future. asking "hey can you tag AI posts?" full stop would have resulted in me doing so with zero comment.

2. disseminating factual information of how gtp works that corrects popular rhetoric based on falsehoods is not a defense of anything, and i think it's pretty concerning that you frame fact checking as inherently oppositional in some way. that's not a healthy way to approach any topic you care about and you'd do well to keep that in check.

3. intellectual dishonesty aside (lol), at the very least arguing against something without actually understanding how it works, and spreading unsourced misinformation is the best way to delegitimize your position. there's plenty of actual problems to talk about with regards to chatgpt (the labour rights issue around content moderation is just one example), there's absolutely no need to invent new ones, especially not if they're gonna, to borrow your own turn of phrase, "deepthroat copyright boots" lol.

have a wonderful rest of your weekend.

#cannot overstate how much i do not care about chat gtp. #anon #reply

[astraltrickster]

Thing is, AI is a nuanced topic. Most big tech advances are. I think the only one I've EVER been super close to that was almost 100% cut-and-dry was NFTs and the "metaverse" sucking ass, and even THAT...well, the former at least originated from an extremely poorly thought-out implementation of a potentially good idea (i.e., an anti-attribution loss registry to counter the damage of unsourced reposting, since convincing people outside of artist communities that YES that shit is rude as fuck BECAUSE it has the potential to be actually damaging is like talking to a wall).

There are good uses of it. There are horrible uses of it. We made the necessary breakthroughs in calculation at one of the WORST possible times, socioeconomically speaking. That doesn't mean we can unmake them. It's putting nearly every single common ethics violation in the world of tech as a whole into the spotlight, from the lack of protections against datamining to the abuse of outsourced labor in moderation work and much more. It didn't invent any of them, and in fact is BETTER about some of them than many other things we just let slide (and shouldn't). Technology marches on, in this form or any other. It changes industries and workflows. It CAN be for the better. The BOSSES always push to make it for cheaper and worse. This isn't the first time it's happened - not even in your lifetime if you didn't have to lie about your age to get here - and it will not be the last. It's just the most visible relative to its absolutely horrendous timing.

I'm in agreement with...some of the more technically-oriented circles I work with irl that it can be genuinely useful. I PERSONALLY do some hobby work in trying to make it more human-positive, for both accessibility and safety, as well as where that overlaps with just...fun. I'm in agreement with both the popular attitude among actors and commercial artists that the way corporations want to use it is absolutely fucking disgusting, and the recognition that a lot of people out there (especially on the fucking bird site) are being total fucking assholes with it. I'm in agreement with labor organizers that we need to place contractual and cultural limits on how it can and should be used. It's a complex subject!

Because of that complexity, I don't mind AT ALL if someone comes to a different conclusion about using it for small-time funsies than I do.

I DO mind when people resist the shitty usage via reactionary sentiment ("this tech is LAZY and DEGENERATE and ANYONE who uses it for ANY reason is a GREEDY ENTITLED SMOOTHBRAIN PARASITE!" - let's play "count the dogwhistles" btw!), blatant ableism ("there's NO ONE in the WORLD who's SO disabled they ACTUALLY need this to help with anything from basic communication to self-expression, or if there are it doesn't COUNT because the COMPUTER did it, not THEM!"), misinformation about how it works (e.g., the copy-paste myth - if that was how it worked ChatGPT papers wouldn't be absolutely full of fictitious citations and AI hands wouldnt look Like That, you realize this, right?), missing the forest for the trees (please do call out asshole usage of AI but not to the point of just forgetting that people were stealing other artists' WIPs to "finish" and making hoaxes and deepfakes and lean-staffing and cheating people out of credits LONG before we had this way to do it and they WILL continue to do so with OTHER new tech as well, an asshole will be an asshole no matter what tools they have; fighting the tech instead of the assholery will only let it happen again as soon as the method changes), using frameworks to fight that asshole usage that are ACTIVELY counterproductive (if it's legally declared that using your work as less than ONE BILLIONTH of my starting point, or developing a style too similar to yours, is infringing on your copyright, this will have WAY more negative impacts than positive; I sure hope the author of your favorite thing isn't the type to go full Anne Rice or JKR; plus even when ignoring the assistive usage of AI as we very much should not, GPT detectors throw a lot of false positives at non-native English speakers, so crackdowns against its usage end up hurting a lot of very vulnerable people far more than they hurt the people using it to do actual tangible harm), or just plain using it as an excuse to bully random other people who individually have about as much power to impact The Issue as dropping a soggy pea on the floor does.

...I also mind when people go full There Is No War In Ba Sing Se about it, but that's not a problem I encounter as much here on tumblr.

#tech #ai tag #current events #serious tag

[venus-is-in-bloom]

i do have more thoughts on the state of writing in media consumption, though. like... with this recent panic about ai art depreciating the value of the drawn image, i've been thinking about how there really is a sort of mystique that clings to most forms of art. like, when someone points at a painting and says, "anyone could do that" or "i could do that", it's meant to insult the supposed artistic merit of the painting or the skill that went into it. there's a default presumption that you need to be a special kind of person with unique, mysterious talents to produce worthwhile paintings. if anyone could do it, it has no value—and ai art threatens very much to make it so "anyone can do it" for a much wider range of images.

but for the written word, "anyone could do that" is just plain fact. creating a perfect imitation of my writing, or toni morrison or ted chiang or anyone else's writing, is merely an exercise in copying.

then there's all this stuff about supposedly developing a unique style that you then market to people in the hopes of getting picked up as an illustrator, which with the recent panic around ai art has gone so far that some people are throwing around the idea that your unique style could be attached to a copyright—implying a disturbing world where every symbolic body, every association, and every concept is parcelled off like land until nothing is left in common.

after the death of douglas adams, eoin colfer wrote "and another thing", a careful pastiche of adams' books in his style, as a tribute to him. an entire (extremely racist!) genre has sprung up in imitation of j.r.r. tolkien's middle earth works, which he spent his entire life developing, and another (often racist!) genre exists purely to mimic the aesthetic and stylistic tricks of h.p. lovecraft's eugenicist horror. in all art forms, pioneers produce the roots on which entire movements grow... it is only in the matter of writing, however, that it is currently nakedly obvious that there is no difference between inspiration and imitation, and that if there is something truly exclusive or scarce about a writer's "talent", it's certainly not physical enough to be enshrined in law. writing fiction is not a rare or precious power, and that's why only the very wealthy can make any money off it.

it's for this reason that things like chatgpt have not received nearly the amount of negative attention that the so-called "ai art" engines have—only a few people piggybacking off the anti-visual-art side of the movement have attempted to invent a hitherto unheard-of preciousness for writing to possess. for the rest of us, the illusion that fiction writing is prestigious, mysterious, or market-scarce remains faint and unconvincing—we cannot convince ourselves that we'll one day get rich off our beautiful talents. we can only do it for love of the art and its natural powers.

#things venus said