Home
Artificial Intelligence
We have solved the first of the three fundamental language challenges in Artificial Intelligence. Remi’s language algorithm represents a major milestone for general AI that can understand and communicate with us.
Remi has learnt how to talk.
Remi is an exciting breakthrough in Artificial General Intelligence. Through a new type of reinforcement learning, our artificial intelligence has learnt how to talk through conversing with another AI. Meaning that she can understand what you’re asking. (At this point she’s only learnt about 20 subjects, but she’s improving.)
Remi’s knowledge bank is growing.
Remi is connected to a wide range of services and information streams. We’re trying to teach her as much about the world as possible. From wine varietals to hotels to the stock market we’re teaching her new things every day.
One big brain.
We were determined to build one singular AI. This means Remi is capable of handling complex requests easily. Things like ‘Can you get me a taxi to my meeting with John Smith and find me a good coffee on the way’ are simple for Remi.
Intelligence is a beautiful thing.
A beautiful artificial intelligence platform
Remi examples
“Remi, can you grab me a taxi to my meeting with Julia and find me a cafe near her office?”
“Remi, what’s the cheapest weekend to stay in Barcelona in November?”
“Remi, has the Tesla stock price ever surpassed $250?”
Remi, is it likely to rain during my trip to Melbourne?”
“Remi, can you order a margherita pizza for when I arrive home and find me a wine to suit on the way?”
How it works.
Remi is built on the world’s first successful general language reinforcement learning algorithm. An extremely exciting breakthrough in AI research. This algorithm teaches itself how to converse and to understand the underlying meaning and associations through trial and error. Just like humans do.
We’ve achieved this by setting up two AI agents in an environment where they need to work together to complete a task. Their only means of interaction is to talk to each other. In this environment, Remi learnt how to converse and work together.
Remi is connected to a wealth of popular services and products so you can integrate Remi with many apps you already use.
Who we are.
We are a bunch of Entrepreneurs, PhDs, Research Scientists and Designers who love research and beautiful products. Our primary mission is to understand intelligence by recreating it. We believe that the human-level artificial intelligence will be one of the most beneficial inventions in human history and we work to use our research in applied products that benefit mankind.
Our present research focus is holistically on language understanding in general reinforcement learning. One of the most fundamental paradigms in Artificial Intelligence.
Our Research
Research at Remi is centred upon Artificial Intelligence that takes inspiration from both cognitive neuroscience and evolutionary theory. We believe that emulation of both areas will be required to achieve human-level artificial intelligence.
Our short-term research focus is on language acquisition, conceptualisation in the hippocampus and the formation of long-term memories.
Contact Us
Press: media@remi.ai
Careers join@remi.ai