Picture: for illustration purposes

Google Levels-Up Battle With Microsoft with Advanced AI Offerings

Published September 21, 2023
8 months ago

Google has unveiled an array of new artificial intelligence technologies and strategic partnerships aimed at making a more significant inroad into businesses and enterprises. The tech giant made the announcement at their highly-anticipated Google Next conference in San Francisco, marking a crucial step in catching up with their rival, Microsoft, on the AI front.

Among the new introductions were high-profile clients for its cloud software - General Motors and Estee Lauder Companies - a clear indication of Google's resolute drive into the enterprise cloud market. Google introduced new AI chips built specifically for enterprises, a tool to watermark and track AI-generated images, along with advanced security tools and improvements for its office suite.

As Microsoft has already embarked on an ambitious AI strategy, Google's transformative moves serve to bolster its footing in the competitive AI landscape. Acknowledging the distinctive pace of enterprise customers, Google Cloud CEO Thomas Kurian advises a slow and methodical approach for businesses integrating AI into their strategic developmental journey.

To amplify Google's enterprise cloud service, it unveiled 20 new AI models, increasing its collection to a commendable 100. In momentous collaborations, Google Cloud customers gain access to Meta Platforms' AI model LLaMa 2 and start-up Anthropic's Claude 2.

In a landmark initiative, Google's new edition of its foundational AI infrastructure sees performance improvements and additional functionalities. Its new version of the text model, PaLM, enables users to input larger lengths of text, easing the process for lengthier documents.

SynthID, a tool to watermark AI-generated images subtly without human detection, also featured among the new offerings, providing a fantastic tool for ensuring the integrity of images in the digital space.

Google also took a stride forward in database management, with a new AI-driven tool capable of transferring databases from Oracle to an open-source version, simplifying a historically challenging task.

The announcement also highlighted an optimised version of Google's fifth-generation tensor processing unit (TPU), called TPU v5e, designed to train large models and serve content efficiently. While it doesn't match the power of Google's soon-to-be-launched flagship AI chip, customers can access this 'supercomputer'-equivalent TPU to tackle more complex computing problems.

Leave a Comment

Rate this article:

Please enter email address.
Looks good!
Please enter your name.
Looks good!
Please enter a message.
Looks good!
Please check re-captcha.
Looks good!
Leave the first review