Artificial intelligence is gaining momentum and is being increasingly used across industries, but more importantly, the lines between artificial and human intelligence are getting blurred. Google is blurring them even further by endowing artificial intelligence with imagination.
Google owned AI lab and DeepMind is working on endowing AI with imagination which would open vast possibilities for the technology — AI would be able to reason through decisions, make plans for the future, and even dream.
AI isn’t replacing humans just yet, and it might never be able to even, given the fact that it is based on algorithms. The only problem with Google AI is that it will use algorithms that can easily adapt to changing conditions, even those it hasn’t been programmed for.
On its website, DeepMind explained how imagination in the context of AI will work. It said: “When placing a glass on the edge of a table, for example, we will likely pause to consider how stable it is and whether it might fall. On the basis of that imagined consequence, we might readjust the glass to prevent it from falling and breaking. This form of deliberative reasoning is essential ‘imagination’, it is a distinctly human ability and is a crucial tool in our everyday lives.”
AI isn’t getting human skills such as reasoning or critical thinking, but it is developing a deeper understanding of consequences and results. It is developing a skill set to navigate complex situations and increase adaptability. It is getting the ability to look towards different outcomes of an action and finding out which action would provide the best outcome in a particular situation.
What’s important to note is that it is using the same basic skills that humans use. It is using a combination of reinforcement learning which is learning through trial and error method and deep learning — learning by processing large amounts of data in a way similar to the human brain.
DeepMind is nowhere near replicating human like intelligence, but it is creating a model that can be scaled up and down, depending on the situation.
"Further analysis and consideration is required to provide scalable solutions to rich model-based agents that can use their imaginations to reason about – and plan – for the future,” the company says further on its website.
Endowing bots with the capability to think before they act, might not necessarily be a bad thing, it might actually minimize risks involved with the technology. In a scenario like the one predicted in the 2004 science fiction movie I, Robot, this would prevent robots from initiating anarchy. The world is a complex place and AI is just beginning to navigate it.
That being said, we are actually crossing a new barrier in AI, we are making it stronger and are opening the path to it getting human-like capabilities.