The Lamen

AI will serve as one of the building blocks for Google's new search engine.

A NEW SEARCH ENGINE IS IN THE WORKS AT GOOGLE

by | Apr 17, 2023

i

Artificial Intelligence.

GOOGLE IS GOING FULL FORCE INTO THE AI RACE BY BUILDING A NEW SEARCH ENGINE FROM THE GROUND UP.

The announcement of ChatGPT by OpenAI triggered a “code red” at Google in just a couple of weeks. It’s been a few months since then, and Google is now looking to make some radical changes to win the AI race — by building a new search engine.

The first serious threat to Google’s search engine in years, Microsoft’s new AI-powered Bing has quickly made the industry shift its focus to improving search and workflow with these large language models (LLMs).

  • Google unveiled its AI chatbot Bard back in February as “an important next step” in the company’s AI journey. The announcement came after Google created an AI-specific task force about two weeks after the ChatGPT demonstration.
  • The company also invested $300 million in Anthropic, an artificial intelligence company founded by ex-OpenAI employees in February last year. The company has since then launched its own AI chatbot called Claude.
  • The rushed release of Bard has led a shitstorm of its own, with the chatbot being called out for making a factual error in its very first demo, along with reports that Google is forcing its DeepMind division to train Bard using ChatGPT’s data – leading to the resignation of a prominent AI researcher.

According to a new report from the New York Times, Google is speeding ahead to build a new search engine powered by AI chatbot technology, while adding an array of AI features to the existing one.

  • Codenamed “Project Magi,” Google currently has 160 designers, engineers, and executives developing and continuously tweaking these new features.
  • Currently planned additions include the ability to answer questions about software coding or write code based on a user’s request, which would potentially use Google PaLM (Pathways Language Model) model.
  • Google has already begun testing these features, with the company inviting some employees to test out the search engine’s ability to hold a conversation by asking follow-up questions.
  • These features are initially expected to roll out to a maximum of one million people, which would progressively grow to 30 million by the end of the year – all exclusive to the United States.

According to the report, these developments have been a result of Google’s fear of missing out on a large revenue source.

A NEW SEARCH ENGINE DOESN’T JUST HAPPEN OVERNIGHT.

These developments come as a result of the Samsung scare back in early March — when the company was considering replacing Google with Microsoft’s Bing as the default search engine on their devices.

  • While no clear reasons were given, Samsung reportedly considered this change thanks to the lead Microsoft currently exercises in terms of their search engine’s AI capabilities.
  • Google would be losing out an estimated $3 billion in annual revenue on the Samsung contract – explaining the state of “panic” for a company that runs 90 percent of internet searches and nearly 70 percent of smartphones.
  • As a result, Google is looking to build a new search engine that would offer “a far more personalized experience than the company’s current service,” perhaps putting together a pitch to avoid losing business.
  • However, this new search engine is still in its early stages, and there’s no clear indication of when it will release.

Google may have been reluctant in deploying the new search engine technology, especially since the chatbot is not the best way to serve users digital ads.

  • “If Google gives you the perfect answer to each query, you won’t click on any ads,” Arm Awadallah told the NY Times.
  • Search ads remain Google’s top money maker, bringing in nearly $160 billion, or nearly 60 percent of the company’s revenue. As such, the company plans to keep ads in the mix with Magi, as the NY Times reports that Google may still place an ad underneath the answers.

Big tech companies continue to release AI-infused products – even if they come with their own range of consequences of privacy concerns, hate speech, and the simple fact that they may spew nonsense.

  • A chatbot built for the medical world, Google unveiled the Med-PaLM 2 last month – which performed at an “expert” doctor level, scoring 85 percent on medical exam questions, the company stated.
  • Google also announced new AI features for its Workspace apps, including Gmail, Docs, Sheets, and Slides – with features like proofreading, auto-generating images, and automatically capturing notes in Meet.

People have varying beliefs about Google’s rollout of AI products – with some finding it too slow, others just hasty. However, the industry is surely undergoing rapid turbulence – without indication of who comes out on top.