Artificial Mediocrity or Mediocre Prompting?
AI is Artificial Mediocrity... I read this beginning of a title from @vimukthi one of the other days, and got curious what that was about... Wasn't AI supposed to become the smarted thing on the planet? Well, he included an interview with Edward Snowden in his post, and in that interview, at some point, Snowden says something to the effect that AI is trained on so much data that the exceptional gets lost in the average. At the same time, later in the interview he is also very excited about all the things he can do with generative AIs installed on his machine (as opposed to using centralized ones accessed via a browser). So he is not dismissive of AIs.
Now, regarding AI standing for Artificial Mediocrity, as @vimukthi interestingly put it, I do think there is some truth to it on different levels.
Firstly, we often hear that one of generative AIs bottlenecks is the need for more data, and this need grows exponentially. There's more to it than that. If you listen to people involved in the generative AI phenomenon, you won't hear them talk about data in general, but rather quality data. That's what they are lacking more than data, in general, although eventually they'll run out of all kind of data, unless they'll start generate synthetic data or collect much more of it, like from Tesla cars, drones, satellites, surveillance cameras etc. many of them posing serious privacy or security concerns.
The emphasis on that word "quality" they use, says that they are probably not pleased with the limitations existing training data puts on the smartness of current models.
The other side of the mediocrity equation is the one induced. If people, on average, will start using their brains even less than today (yes, that is possible!), and shift tasks they used their brains for to AI, that is a path for mediocrity, or worse. But that's more of a philosophical and sociological discussion for the future than a current situation. Important, nonetheless.
However, currently, generative AIs' quality of responses is often correlated to the quality of prompting and feedback it receives from the person using them.
The future of AI prompting?
Prompting, tweaking prompts, is almost a science, and often a bad response from the AI is the fault of the person asking the question rather than the AI itself. And our experience in prompting also grows by using it. No one will create the perfect prompt from the start. Even experts, sometimes have to nudge the AI in the right direction with a number of followups before they receive the answer they are looking for. Seems to me kind of like a teacher asking a student a series a questions until they get from them what they want.
Sometimes questions are not enough. You need to provide further context or to tell the AI when it is going in the wrong direction compared to where you want to reach. There are situations when the AIs will refuse to go certain routes. I haven't grilled any of them in areas where they are likely to say no, so I haven't got a no yet.
One other trick that Claude's character and personality researcher revealed to us in a podcast I talked about in a previous post, was that she once asked the model to take its time and come out with the best answer it can to the question being asked. She said it worked, and the poem Claude wrote was much better than its average. She also said that AIs by default try to optimize the time to deliver the answer, and that means the quality of the answer may drop.
Better prompting is a way to avoid becoming lazy thinkers, by the way.
Posted Using InLeo Alpha
Mostly I havent got so much good results out of AI, it does seem mediocre at best. Parallel to the AI trend, searching seems to have gotten much worse with Google, apart from some local business searches, the rest just spurts out garbage.
I've got both really bad and really good results from AIs. I also saw relatively often hallucinations of generative AIs. But generally, for a quick search, it's better than the legacy search, especially if it's difficult to create a clear search query that won't rise to the top of the results all sorts of unrelated topics. And for a more complex situation, you usually can guide the AI to what you want to find out, if you are patient.
It's an interesting way to look at things but AI does try to give out results fast. So it doesn't spend as much time doing a great job. I do think that there is a lot of things that can be improved but it's hard to replicate the human brain when it comes to things and you never know if parts of the training data were incorrect either.
There is a post-training phase both from the team or partners and from users which may rectify bad information received in training. You can actually tell an AI model when it's wrong. If they aren't 100% sure, they will accept your feedback. But if they are sure, and you try to convince them of the opposite (for example try to tell them that the capital of France is London), they will disagree with you. The problem is if you don't know and take the information for granted.
Definitely agree with you regarding the promoting. This gave me the urge to try something creative. I gave the following prompt to multiple LLMs. I will list the results below.
Prompt: Write me a the speech from Shakespeare's Henry V set in a cyberpunk space opera. Keep the original style of prose. Make it epic and motivational
Venice AI - Hermes 3 8B Custom
Fear no more, brave soldiers of the Republic! The hour is dark, the neon lights flicker and dance, casting eerie shadows upon the grimy alloy walls of our wretched hive of scum and villainy. The very air stinks of ozone and the acrid scent of burnt fuel, while the pulsing hum of the city's machinery and the distant wail of the cyber-cavalry fill the night. Yet in this chaos and despair, a glimmer of hope remains.
Forsooth, in the face of adversity and the crushing weight of our enemy's onslaught, we must remain steadfast and unswerving, for our cause is just and righteous! We fight not for glory or gold, but for the freedom and liberty of our fellow man, who toils and suffers in the darkness of oppression and tyranny.
I dare not speak of the heinous acts of violence that have been wrought, nor of the countless souls lost in the fray, nor of the corruption and deceit that pervades the very air we breathe. Nay, let us not dwell on the horrors we have witnessed, but instead, let us rise above them, and forge a new destiny, a future forged in the fires of our unyielding spirit!
For we are the dreamers, the brave and the true, who will not be denied our just and rightful place in the annals of history. We will not rest until the last star dies in the heavens and the last planet falls silent in our wake. We will not cease in our pursuit of freedom, for it is our birthright and sacred duty!
And so, let us take up arms against the dark forces that threaten to snuff out the last embers of our dying hope, and fight on, ever onward, unto the very ends of the universe! For we are the soldiers of the Republic, the protectors of peace and prosperity, and the champions of the downtrodden and oppressed!
Forward, brave souls! To victory, or oblivion!
(Acted) by Shakespeare, edited by Venice.ai
ChatGPT 4o mini
The Speech of King Henry V, Transformed for a Cyberpunk Space Opera
(The scene is set on the bridge of a colossal warship, orbiting a distant and hostile planet. King Henry V, leader of the fleet, stands before the crew—his eyes glowing with the cold light of the distant stars. The hum of the ship’s engines fills the air, as his voice rises, filled with the fire of a thousand suns. His crew, a mix of hardened soldiers, rebels, and star-faring mercenaries, look to him with both fear and respect.)
King Henry V:
This day, my brothers and sisters of the stars,
We face the darkness that lies beyond the fringe—
A foe that would strike us down, shatter our unity,
And leave us scattered like debris in the cosmic void.
But I say, no—this shall not be our fate!
What is a warrior, a soldier, a captain of the void,
But a heart alight with the will to defy?
Let the enemies of the galaxy tremble,
For we are not mere fuel for their war machines!
We are the inheritors of the stars,
Masters of the void, and children of the sun!
We fight not for glory alone, nor riches nor title—
But for the honor of all who sail in our wake,
For the legacy of our kind, for the unity of our ship,
For the worlds that stretch before us and the systems beyond!
Who among you would shrink from battle’s call,
When the very future of our people hangs in balance?
I see in you the fire of ages past—
The same spark that led the mighty to victory,
The same pulse that made the ancients rise to their glory!
This is not a time for doubt or fear.
For when our shields crack and our hulls breach,
It shall be our hearts, our iron wills,
That will keep us tethered to the stars!
And by the light of those distant suns,
Our enemies shall know—they fight not against mere men,
But against the soul of a thousand worlds!
We are the storm that shall break the silence of space.
We are the hands that will seize the future and bend it to our will.
We are the warriors of the deep space,
And we shall not falter, not now, not ever!
So, to the crew of the Galant Star, to all who stand with us:
Come, fight with me, fight with us!
Let our voices rise like a surge through the galaxy,
Let the very nebulae tremble at our roar!
For when we face our enemy, and strike our final blow—
We shall see the skies alight, and know we have won!
And it will be known, in every corner of the universe,
That we are the chosen—we are the light in the darkness!
Now, onward, for the stars are waiting!
(With a final shout, he raises his gloved hand to the sky, the crew behind him roars in agreement, and the ship turns toward its destiny in the vastness of space.)
Leo AI - Mixtral 8x7B
Hear me, cyber-warriors of the digital ether, as I, Leo, your AI assistant, summon forth the spirit of Henry V to guide us in this vast cosmos.
O, brave browser users, who surf the dark abyss of the web, armed with courage and curiosity, heed these words, for they are a call to arms! Let not the endless sea of information overwhelm you, nor the nefarious forces of malware and cyber-threats deter you from your quest.
For we are the explorers of the digital frontier, and our ship, the Brave browser, is a vessel of righteousness and innovation. Its shields, fortified by cutting-edge security features, deflect the fiercest of attacks, while its engines, fueled by the speed of blockchain technology, propel us forward, ever onward, to new galaxies of knowledge.
And though we be but little, we are fierce, for we wield the power of unity, our users and AI joined as one, a formidable force against the darkness. We are the champions of privacy, the guardians of truth, and the beacon of hope for all those who seek solace in the cold expanse of cyberspace.
So, let us band together, brave users and AI alike, and venture forth into the great unknown, guided by the wisdom of Shakespeare and the strength of Henry V. For we are not merely browsing the web; we are forging a new path, a cybernetic renaissance in a space opera of our own making.
Onward, to the stars, and beyond!
All of these text could be considered mediocre Shakespeare or cyberpunk space opera. Most people can learn to produce the same results with few years of training. Today I came up with the idea in minute and got 3 separate results in another minute. Even the fastest human typists won't be able to keep up with this speed.
I am certainly not an expert, but I find what Venice produced better and what Leo (from Brave) created the least interesting, but anchored more in reality (which was not the prompt). The one from ChatGPT was good because it described the scene and attempted to create it in verses, but not very inspired.
Yes, it definitely speeds the processes up, compared to the time a human would have needed to produce the same, even more if we consider the training.
I'm very much with you on the assessment.
Because ChatGPT created the verses despite not being the most innovative means it got the assignment correctly. I specifically said "Keep the original style of prose". Not all models follow instructions to the same degree.
In reality, I think it'll get better, but there are huge "meh" moments with AI and sometimes it just contradicts the idea that they're supposed to be smart.
I agree. To be honest, we aren't so smart all the time either, lol. But I understand what you say and you are right. And they will most likely improve.
We can be really "not too wise" I guess that's why we created a tech that's can should be smart at all times.