Google Assistant to shift towards Generative AI Model: Aiming to outperform competitors with advanced contextual responses

Google Assistant is reportedly shifting towards a generative AI model using Large Language Model (LLM) technology, according to an internal email leaked by Axios. The Assistant team is restructuring to investigate the possibilities of a more advanced Assistant with generative capabilities, akin to the technology used in ChatGPT and Google's own Bard chatbot. This change indicates Google's awareness of its current AI model's constraints and its attempt to match competitors who have successfully implemented generative AI.

Previously, digital assistants like Google's Assistant, Amazon Alexa, and Siri relied on a Mad Libs-style approach, requiring users to input specific subjects and verbs. The generative model for Assistant could provide more contextually relevant responses using information from the web, similar to the Bard chatbot. However, it's unclear if users want more extended interaction with navigation systems or prefer straightforward queries, and Google's challenge will be to find a balance between generative capabilities and traditional AI interfaces.

As the company prepares for this shift in the coming months, it demonstrates its dedication to exploring the potential of generative AI for Assistant and adapting to future technological advances and user preferences. Jennifer Rodstrom, a Google spokesperson, emphasized the role of LLMs in improving the Assistant's functionality. However, the timeline for incorporating this technology into Google's smart home products is unclear, and the privacy implications of using such advanced AI technology in smart home products raise concerns.