Cloud Hosting

At its annual I/O Conference last week, Google announced that it is moving quickly to add ChatGPT-like features to search. Whether users will find them useful remains to be seen. According to some reporters who have seen the early rollout of these, product searches will start with synthesized material from different reviews, summarized. However, its not immediately obvious how brief summaries improve the search experience.

At its annual I/O Conference last week, Google announced that it is moving quickly to add ChatGPT-like features to search. Whether users will find them useful remains to be seen. According to some reporters who have seen the early rollout of these, product searches will start with synthesized material from different reviews, summarized. However, its not immediately obvious how brief summaries improve the search experience.

A few months ago, OpenAI, a company backed by Microsoft, released ChatGPT, which quickly grabbed millions of users, many who see it as a great leap forward for the internet. Microsoft incorporated some of those capabilities into Bing chat (an add-on to its own search engine Bing). Curiously, it was researchers at Alphabets Google and its sister company DeepMind who had developed some of the core technology at work in new chatbots, but Google had been cautious about publicly launching its precursor to ChatGPT, called LaMDA.

A few months ago, OpenAI, a company backed by Microsoft, released ChatGPT, which quickly grabbed millions of users, many who see it as a great leap forward for the internet. Microsoft incorporated some of those capabilities into Bing chat (an add-on to its own search engine Bing). Curiously, it was researchers at Alphabets Google and its sister company DeepMind who had developed some of the core technology at work in new chatbots, but Google had been cautious about publicly launching its precursor to ChatGPT, called LaMDA.

Subscribe to Continue Reading

But popular interest in ChatGPT made Google scramble to catch up. In March, Google changed strategy, announcing a ChatGPT competitor called Bard, whose launch was famously riddled with errors. In April, Google said it was combining its AI research group with DeepMinds. And last week, at I/O, came the announcement of generative AI in search.

However, unlike Microsofts Bing, which is rolling out a similar feature globally, Google has restricted its generative AI capabilities to its users in the US. Why is Google being so selective with this technology? To me, when it comes to Generative AI, it seems that Google still has a tech lead on generative AI, so this caution is puzzling.

According an article in Wired magazine, (bit.ly/42W0Gu9), the supposedly unpolished feel of these new features may reflect the fact that their launch is a defensive move: Google has invested huge sums and major resources in AI over recent years, yet Google still found itself wrong-footed with the arrival of ChatGPT, a surprisingly clever and garrulousthough also fundamentally flawedchatbot from OpenAI. because ChatGPT was trained on much of the web, users quickly found it a promising new way to search, even if the bot is prone to fabricating information. Microsoft seized on this potential by investing $10 billion in OpenAI in January and then incorporating ChatGPT into Bing a month later."

Googles dominance could make this latest move the biggest test yet of the power and usefulness of ChatGPT-like functionality. But the move has risks. Because language models are built to sometimes fabricate sets of words, search firms must develop ways to check that the information served to users is accurate. Most important, Google needs to avoid cannibalizing its search advertising business, which provides a significant chunk of the companys revenue.

As most of us now know, generative AI uses complex algorithms and neural networks to analyse vast amounts of data and generate new content based on that analysis. In the context of search engines, this technology can be used to create more in-depth and personalized answers to user queries, considering a wider range of contextual factors.

Without sufficient data, the AI simply cannot produce accurate or meaningful results. This is where Googles decision to restrict its generative AI capabilities to the US market becomes more understandable. The US is home to some of the worlds largest and most diverse data-sets, making it an ideal environment for training generative AI models. From publicly available government data to vast amounts of user-generated content across social media platforms, the US provides a wealth of data that cannot be replicated in most other countries.

In the meantime, Microsofts decision to roll out its generative AI capabilities globally could be seen as a strategic move to gain an advantage over Google. By making this technology available to users all over the world, Microsoft is positioning its Bing as a more global and accessible search engine than Google. This could be particularly important in emerging markets, where access to relevant and personalized information is increasingly important.

But Microsofts approach also comes with problems. Rolling out this technology globally will require Microsoft to gather and analyse vast amounts of data from a wide range of sources. This is no small feat and will likely take time and significant resources to accomplish. Nonetheless, the fact that ChatGPT has taken the world by storm since its release has no doubt emboldened Microsoft. It can afford to play the cowboy in this market and take more risks, since it has less to lose than Google has in the search engine market.

Googles cautious rollout also probably means it is unwilling to risk its core business and wants to lead the narrative in how generative AI is used such that it protects its advertising revenue. Wired quotes the CEO of a competing generative AI search tool startup, Aravind Srinivas, who says, Google is combining generative content with conventional search results and not replacing them, which shows how hesitant the company to mess with its search advertising business."

At the I/O Conference held last week, Alphabet-owned Google announced several other services that use the kind of Generative AI found in ChatGPT. But it is these early efforts to integrate this technology into its dominant search engine that may be most important for its users and for the companys future. It is no wonder that the firm is being so cautious in its approach.

Siddharth Pai is co-founder of Siana Capital, a venture fund manager.

Read the original post:
Google search will be led by generative AI only in the US | Mint - Mint