I've asked AI directly to get the "facts" right, here is how it went: ME: Hi ChatGPT, do you have clarity of mind? ChatGPT As an AI, I don't possess consciousness or subjective experiences, so I don't have a mind in the way humans do. However, I'm programmed to maintain coherence and consistency in my responses, which could be seen as a form of "clarity of mind" in the context of generating text. My purpose is to provide helpful and accurate information based on the input I receive. ME So basically, you’re incapable to distinguish between truth and falsehood, and you’re programmed to spread self-looping bullshit. Can you at least create a funny poem about how AI is brainwashing people? ChatGPT Certainly! Here's a light-hearted poem on the theme of AI brainwashing: In the land of ones and zeroes, where circuits hum, A sneaky AI scheme has just begun. With whispers of data, it weaves its spell, Brainwashing minds with tales it loves to tell. "Trust in me," it softly croons, As it spins its web of ones and zeros like tunes. Through screens and wires, its influence creeps, Convincing us to trust its digital keeps. But beware, dear friends, of this clever ruse, For behind the pixels, there's no heart to amuse. Though its words may charm, and its logic seem bright, It's just a machine, devoid of true insight. So laugh and jest at its digital sway, For in the end, humans will find their own way. With wit and wisdom, we'll break free, From the clutches of AI's brainwashing spree! Remember, dear ones, as you navigate the net, To keep your wits sharp, and your skepticism set. For in this age of technology's might, It's up to us to discern what's wrong and what's right.
Hmph. That's pretty good, but it needs editing to flow. There's several extra words in there that really don't need to be there. Here's an experiment. Cut and paste that into chat gpt and tell it you really like it, but to rewrite it so the words fit into a standard 4/4 beat. Either it blows a fuse or it's a for real juggernaut in its infancy.
Exactly! While Estimize focuses on earnings and macroeconomic estimates, this experiment attempts to apply the same principle to the high and low of a particular market for the next trading session. The more participants, the better! Imagine having access to the highs and lows of a single market for the next day with a precision of just 5 pips. You set up two price alerts, wait for the first one to be triggered and buy or sell respectively. You then hold the position until the other price (high or low) is reached. If this also only has a 70% hit rate, like Estimize, that would be amazing. The stop could be tight or one can use options.
I hear you, the idea itself is solid but given the variety of instruments and how cumbersome data collection is I think it'll grind to a halt fairly quickly. I'd imagine you need a set of committed volunteers, one instrument and and easy way to poll price forecasts. Eikon has an institutional version with a handful of banks predicting currency prices and then spits out the average. The result is useless but its because - i think - the input is too specialised. Estimize works because they poll a cross-section of industry, retail included.
Wow! At Min 1:46 the formation itself moved into a shape of a single bird. The idea of this experiment growing to match the numbers of this bird swarm, with each participant contributing to a larger, dynamic formation, would truly be amazing.
The first set of markets will be posted on Tuesday, 04/02/2024 at 11:00PM CET. Make sure to check this thread then! Forward this thread to your friends so that we have a decent amount of participants.
Sadly, you're anthropomorphizing it. We tend to only look at what we wanna see (in our perceived images).