Google’s parent company Alphabet on Tuesday announced how it is building artificial intelligence across its businesses, including an enhanced Gemini chatbot and improved search capabilities, as it faces increased competition from OpenAI and other rivals. It showed whether there were any.
At the annual I/O developer event in Mountain View, Calif., CEO Sundar Pichai said the company will roll out its AI overview to all users in the U.S. this week after a lengthy public test that began last year. Stated.
The new AI capabilities announced Tuesday will help investors gauge Alphabet’s progress as it competes with Microsoft, OpenAI and other competitors for dominance in emerging technologies.
Alphabet’s stock rose during the product presentation and was last up about 1% at $172.50 Tuesday afternoon.
“We are in the very early stages of an AI platform,” Pichai said.
Google announced improvements to its Gemini Pro 1.5 model that can understand large amounts of data.
Google announced on Tuesday that it would double that amount to 2 million tokens. This means that AI could potentially be able to answer questions given the ingestion of thousands of pages of text or over an hour of video.
Pro models start with prompt sizes of up to 1 million tokens or data and are also available to subscribers of Google’s Gemini Advanced service.
In another sign of the fierce competition between OpenAI and Google, online search leader OpenAI has launched Veo, an AI model it claims is the most powerful ever for creating videos with simple text commands. teased.
Google released its previous video generation technology in January, but it was improved upon by OpenAI’s Sora a few weeks later.
ChatGPT makers are touting their movie-producing software among Hollywood executives, captivating and worrying the creative industry.
Google announced that movie director Donald Glover experimented with its AI.
The company also previewed its new text-to-image model, Imagen 3, and touted collaborations with other artists.
The company announced a scaled-down version of Gemini called 1.5 Flash, aimed at lowering AI deployment costs and speeding up response times.
Like its more capable versions, Flash can capture large amounts of data while being optimized for chat applications, video, and image captioning.
AI Oververs uses generative AI to synthesize information and answer more complex queries for which there are no simple answers on the web.
Google DeepMind, Alphabet’s AI division, has been working to build technology that can perform everyday tasks for consumers.
Early results appear in Project Astra, a tool that allows you to draw conclusions about the world around you using your smartphone’s camera.
In a demo video shown during Google I/O, users deployed it to identify speakers and find glasses left somewhere else in the room.
Microsoft-backed OpenAI on Monday unveiled a new AI model called GPT-4o. This allows ChatGPT to respond and be interrupted with voice in real time. Both of these are characteristics of realistic voice conversations that AI voice assistants like Google Assistant have found difficult.
