Under intense pressure to compete with ChatGPT — the much-talked-about AI chatbot that has become a viral sensation — Google announced Monday that it will release its own “experimental conversational AI” tool called “Bard.” The company also says it will add new AI-powered features to Google Search.

Google will first grant access to Bard to a group of trusted external partners, according to the company’s blog post on Monday. It said it plans to provide public access “in the coming weeks.” Starting this week, general users will have access to search results that display AI-generated text, especially for complex queries.

Google has been using AI to power its products behind the scenes for years, but has never released a public version of its conversational chat product. ChatGPT, an AI conversation tool that can auto-generate essays, poems and even entire movie scripts created by his startup OpenAI, has gained 100 million users in just two months of its launch. Did. Do this move. Google’s announcement comes a day before Microsoft is expected to announce details of plans to integrate ChatGPT into its search product, Bing (Microsoft recently invested ChatGPT’s creator OpenAI in his $10 billion was invested).

Since the launch of ChatGPT, Google has faced tremendous pressure to bring its AI technology to the public more. Like other big tech companies, Google has delayed technological breakthroughs similar to previous inventions like Search, Maps and Gmail, betting that its next big innovation will be powered by AI. However, the company has historically kept his AI work potential completely hidden, especially with conversational AI tools, only allowing Google employees to test chatbots internally. This release shows that fierce competition has pushed Google to put its achievements in the spotlight.

“AI is the most important technology we are working on right now,” Google CEO Sundar Pichai wrote in a blog post announcing the change Monday. “That’s why we changed the company’s direction six years before him to align with AI, and our commitment to organizing the world’s information and making it accessible and useful to people around the world.” We see AI as the most important way to fulfill our mission.”

In a Google blog post, Bard, a new AI tool, “seeks to combine the breadth of knowledge from around the world with the power, intelligence and creativity of large-scale language models.” Specifically, it means that NASA’s James can explain new discoveries from his Webb Space Telescope in a way that a 9-year-old can understand. according to the company.

According to the release, other examples the company has given Bard include planning a friend’s baby shower, comparing two Oscar-nominated movies, and getting recipe ideas based on what’s in the fridge. There are some things that can help you.

All these possibilities seem convenient and convenient for the user. However, new technologies tend to come with potential downsides as well. Google is one of the world’s most powerful companies, requiring far more political and technological scrutiny than smaller startups like ChatGPT’s OpenAI. Some industry experts have already warned that tech giants like Google may miss the potential harm of conversational AI tools as they race to compete with OpenAI. And if these risks are left unchecked, they can reinforce negative social stigma and upend certain industries, such as the media. Pichai acknowledges this concern in a blog post.

“It is important to bring these model-rooted experiences to the world in a bold and responsible way,” Pichai wrote.

This may explain why Google initially released its AI conversational technology only to “trusted partners.”

So perhaps your first exposure to Google’s conversational AI technology will be a new search feature that “extracts complex information and multiple perspectives into an easily digestible format.”

For example, when someone searched for a question with no right or wrong answer, such as “Which is easier to learn, piano or guitar, and how much practice is required for each?”, Google said: Use AI to provide nuanced responses. In the sample answer pictured below, he offers two different interpretations of what sounds like an essay or blog post: “Some say… others say.” This isn’t the simple answer you’re used to in Google’s Q&A snippets.

What would Google’s AI-generated search results look like?
courtesy of google

At this point, these announcements seem to be just teasers, and it looks like Google needs to reveal more about its AI capabilities. The real test for Google’s AI tech as it rolls out is how it compares to ChatGPT. ChatGPT is already making headlines and attracting real-world applications, like using BuzzFeed to auto-generate quizzes that job seekers use to write their covers. letter.

Google, a trillion-dollar company that billions of people use every day, is in a tough spot. For the first time in years, the company faces significant challenges from relatively newcomers in one of his core competencies: AI. Generative AI, the kind of AI that powers chatbots, is the most exciting new form of technology in Silicon Valley.

Google built some of the foundations for this technology (the ‘T’ in ChatGPT is named after the tool Google created), but it’s still leading the way in showing the world what this kind of AI is. It’s ChatGPT, not Google, that wins. can. Whether Google can pull off a similarly high profile with this new tool could determine whether the company remains the leader in organizing the world’s information, or whether it cedes that power to new entrants. there is.


Source link