Google unveiled its newest conversational artificial intelligence (AI) model LamDA, which can almost hold its own in normal conversations. The LaMDA, which stands for Language Model for Dialogue Applications, was introduced during the first-day keynote of Google I/O, its yearly developer conference, earlier last week. Google researchers aim to replace artificial, robotic conversations with AI, with more natural dialogues.

The LaMDA was showcased holding two conversations with humans on extremely difficult, even odd subjects such as Pluto and paper aeroplanes to show how the LaMDA could reply to odd, natural questions instead of being stumped when users make generic conversational statements. “Language is remarkably nuanced and adaptable. It can be literal or figurative, flowery or plain, inventive or informational. That versatility makes language one of humanity’s greatest tools — and one of computer science’s most difficult puzzles," the company said in a blog post.

The LamDA is Google's attempt to reduce the learning curve that humans face while interacting with AI chatbots and other conversational software. Currently, conversations with AI are scripted as they respond to commands, not hold conversations, which gives an impression of the narrow and limited scope of interaction. For example, Google Assistant and Alexa lose a conversational thread once in a while.

“That meandering quality can quickly stump modern conversational agents (commonly known as chatbots), which tend to follow narrow, pre-defined paths. But LaMDA — short for 'Language Model for Dialogue Applications' — can engage in a free-flowing way about a seemingly endless number of topics, and ability we think could unlock more natural ways of interacting with technology and entirely new categories of helpful applications," the company said.

AI algorithms such as LaMDA, GPT-3 and BERT, known generically as Natural language processing (NLPs) models were created to ensure that the robotic nature of conversations with technology can be solved. Google's LaMDA is based on Transformer neural networks which were the building bases of NLPs like BERT and GPT-3.

LaMDA will help Google make improvements in their AI-conversational products such as Google Assistant on Android phones and Nest Smart Speakers which are designed to be in-home assistants that work naturally. Currently, to activate the products, one needs to make 'wake' the product for example, “Hey Google", followed by specific commands like “turn off the lights". Systems like LaMDA hopefully make such commands more fluid and natural. They can also enhance chatbots and other automation applications, an industry that saw significant growth during 2020.