With the rise of LLMs and the dominance of OpenAI and alternatives, we felt the need to dive in deeper into use cases, integrations, and ethical as well as business and alignment considerations. As such, we've been piloting several different efforts using LLM engines to test market viability, and to gain hands on experience. Two projects are the result of these experiments: Caimeo and Redpill.
Caimeo is an autonomous agent crisis simulation engine. It can be used to simulate an environment, populate it with predefined actors, and run thousands of simulations to identify the probability of a certain success or failure scenario. For example, a school shooting could be simulated a thousand times and result in tragedy 970 times, indicating a high likelyhood of failure.
Then, a modification can be made: what if the teachers all knew Jiu-jitsu, and there was a safe room at the end of every hallway? The probability of success (no-casualty resolution) might rise to something like 65%, clearly indicating what steps can be taken to mitigate public disasters. The same can be applied to environmental disasters and any other human-affected situation which can have different outcomes based on different personalities, skill sets, and convictions.
Caimeo is coming out with an MVP after its initial fundraise in Q4 of 2024.
Redpill is an API gateway that transforms ChatGPT into "Truth GPT", capable of verifying AI-generated content.
It wraps the ChatGPT API (for now - later on other models will be added) and adds an incentive layer for consensus on the "fakeness" of content. As the redpillers judge content, they get on-chain merit, which increases their voting power as long as they keep using it, creating a layer of truthtelling reputation, but preventing the accumulation of power in first movers by causing merit decay as the power is used less.
We believe Red Pill is essential in today's world of content-of-unknown-origin, and that it will be absolutely necessary in this world of information firehosing.
Utilizing our wealth of experience and contacts in the industry, we feel confident we can provide you with adequate guidance on implementing various LLMs, the advantages and disadvantages of various models, and the best approach for specific frameworks, products, and use cases. The field of AI is a rapidly evolving one and we do our best to stay on top of things - so that you don't have to.
Alignment is a popular concern in the field of AI. There is little we can do to align AI with humanity. There is, however, lots we can do to align your business and your employees with AI. The most common (and well justified!) fear from employees today is - will AI replace me? The most common (and just as well justified!) fear of business owners is - will my competition out-AI me? The answer depends on your alignment and level of forward-thinking attitude. We can help you understand both sides of this argument.
Frankly, the raw truth is that yes, some employees will likely lose their jobs. And yes, AI has the potential to multiply your revenue and/or efficiency. It is important to understand in all this that AI is not a silver bullet - thoughtful integration is necessary, identifying easily automatable pain points, and replacing the most narrow of bottlenecks is what generally produces the best results. We're here to identify these replaceable bottlenecks, and to train your employees to enhance their productivity with AI. As the saying goes, you won't be replaced by AI, but by people who use AI to make themselves more efficient.