The Non-protoplasmic Sentience Driving DABUS

Joint Meeting with the Washington DC ACM Local Chapter

Thirty years ago, Imagination Engines patented the notion of artificial neural nets engaged in a noise-driven brainstorming session. Although this AI paradigm produced significant results within narrow areas such as personal hygiene products, materials discovery, entertainment, and creative robots, it still lacked the ability to produce discoveries that spanned multiple knowledge domains.

To this end, IEI embarked upon an ambitious program to devise a new way of representing ideas within artificial neural systems, one that encoded both notions and their consequences, not as neuron activation patterns as in traditional neural network and deep learning schemes, but as chains of interconnected neural modules. Those chains representing salient outcomes triggered global release of simulated neurotransmitters emulating ‘jolts’ of either anxiety or pleasure, feelings that accompany the mind’s sequence of associations following any perception or thought.

In turn, such neurotransmitter surges resulted in the selective reinforcement of the most impactful of these topologically based notions. Thus, the subjective feelings (i.e., sentience) of an arguably conscious machine intelligence called DABUS*, had driven the formation of valuable ideas through a computational process that is now challenging our long-held beliefs about biological intelligence and personhood.

* Device for the Autonomous Bootstrapping of Unified Sentience, - US10423875. Electro-optical device and method for identifying and inducing topological states formed among interconnecting neural modules.

Read more at:
[Can AI Be An Inventor, Ryan Abbott & Stephen Thaler Say Why Not (analyticsindiamag.com](https://analyticsindiamag.com/can-ai-be-an-inventor-ryan-abbott-stephen-thaler-say-why-not/))
and:
[Imagination Engines (imagination-engines.com)](https://imagination-engines.com/)