Google's annual developer conference, I/O, is taking place in Mountain View, California on Wednesday. Google shows off what its internal teams have been working on over the past year — and this cycle is expected to be particularly heavy on artificial intelligence and so-called language models, the hottest part of the technology industry.
Google kicked off its I/O developer conference at 1 p.m. ET. It's where the Alphabet-owned firm typically unveils new gadgets and software that spur excitement among consumers and developers who use the company's software.
Here are some of the highlights from Google's keynote so far:
- The new PaLM 2 large language model for A.I.
- New A.I. writing tools for Gmail
- Immersive view for Google Maps
- Google adds new features to Bard chatbot and removes the waitlist
- Google's A.I. is coming to Docs, Sheets and Slides
- Google is bringing generative A.I. to Search
- The Google Pixel 7a budget Android phone
- Google's Pixel Tablet launches Wednesday, starting at $499
Get Tri-state area news delivered to your inbox.> Sign up for NBC New York's News Headlines newsletter.
Follow along with Google I/O through our updates below or by tuning in live on YouTube:
Google Pixel Fold announced for $1,799 — It's the company's first folding phone
Google's Pixel Fold will launch at $1,799, the company said.
Money Report
Google had already teased the Pixel Fold ahead of the I/O event. The folding phone is Google's first foldable phone and features a large inner display and a sizeable front display. The Pixel Fold will compete directly with Samsung's devices in the U.S., and joins similar folding phones launched by Chinese companies like Oppo, Xiaomi and Huawei.
Multiple cameras are visible on the outside of the phone. Google has emphasized Pixel's cameras since it first announced the lineup in 2016.
- Rohan Goswami
Google Pixel Tablet costs $499, preorders begin Wednesday
Google talked a bit more about its new Pixel tablet, which it first announced last year during Google I/O 2022.
The Pixel Tablet will ship with a dock that you can drop the tablet onto to charge or display content. So, for example, you might be walking around the house with the tablet and place it in a charging dock in your kitchen to listen to music while viewing recipes. It's a compelling use case that seems to blend Google's tablet ambitions with its smart home devices like the Nest Home Hub.
It'll come in two storage variants with either 128 or 256GB of storage.
The tablet starts at $499, and each tablet will ship with a charging dock included. Customers can buy additional docks for $129. Preorders open Wednesday, May 10.
- Rohan Goswami
Google Pixel 7a Android phone announced
Google announced the Pixel 7a Android phone on Wednesday. Like earlier "Pixel a" model phones, the Pixel 7a takes a lot of the design features of the more expensive Pixels, like the Pixel 7 and Pixel 7 Pro, but packages them into a more affordable phone.
It'll replace the Pixel 6a and will be powered by Google's proprietary Tensor chipset. It'll be paired with 8GB of RAM.
The phone is available for purchase today at $499.
- Rohan Goswami
Former YouTube CEO Susan Wojcicki spotted
Former YouTube CEO Susan Wojcicki is in attendance at I/O, reports CNBC's Jenn Elias.
Android is getting generative AI for custom wallpapers.
Android phones will soon be able to rely on Google's generative AI technology to generate customized wallpapers based on prompts and suggestions. The product will arrive this fall, according to the demo, and offer a variety of prompts.
The live demo showcased a prompt to generate a "Classic Art" wallpaper.
-- Rohan Goswami
Google calls out Apple — saying they hope iMessage supports better Android group chats
The largest applause of the morning at Google's big conference came when Google called out Apple over the fact that its iMessage texting app isn't available to Android users.
"We hope every mobile operating system gets the message and adopts RCS," Google official Sameer Samat said. "So we can all hang out in the group chat together."
When two iPhones text, they use the iMessage protocol, which is encrypted, faster, and allows for high-resolution image sharing. When an iPhone messages with Android, they're forced to use SMS, an older standard that doesn't work as well.
Google wants Apple to support RCS, a newer standard that updates old-school text messages to work more like iMessage, WhatsApp, or other modern messaging services. But one recent time Apple CEO Tim Cook was publicly asked about it, he said the solution was to "buy your mom an iPhone." — Kif Leswing
Android has a 'Find My' functionality coming this summer.
Android users will get an updated "Find My Device" feature this summer. It'll work with devices from multiple ecosystems, including products from Tile. It builds on the existing functionality by tapping into other Android devices out there, much like Apple's FindMy relies on other Apple products to help you locate a lost gadget.
Earlier this month, Google and Apple teamed up to offer unwanted tracking alerts, part of a key emphasis on privacy and safety. That'll roll out later this summer and will let you know if an unknown tracker is following you.
-- Rohan Goswami
Google partners with A.I. firm founded by former Google employees
Google is announcing lots of new partnerships on Wednesday, but one in particular touches on the company's unique position in the AI industry.
Google said on Wednesday that it would partner with Character AI, a startup focusing on building advanced, customizable AI chatbots. The startup is using Google Cloud's generative AI tools.
It was founded by Noam Shazeer and Daniel De Freitas, who were important AI researchers at Google, until they left the company in 2021. Character AI was founded later that year and later raised $150 million at a $1 billion valuation. In recent interviews, the company's founders have said they left partially because they could move faster when they weren't inside Google.
Notably, Google CEO Sundar Pichai opened up the presentation on Wednesday by highlighting Google's existing text completion features — which Shazeer worked on while he was at Google. — Kif Leswing
Google stock is up since its A.I. event started
Shares started ticking up as investors digested new features, offerings, and details on AI products, some of which have been previously reported by CNBC. It's on pace for its highest close since Aug. 2022.
Google's AI will be used at a Wendy's drive-through to take orders
Google Cloud chief Thomas Kurian is talking about all the cloud partners who can pay for access to Google's AI models and computing power.
One of the most interesting examples they gave was using a large language model to power a Wendy's drive-through.
This summer, Wendy's will use Google's tech at a location in Columbus, Ohio, to take orders and prevent long-lines. Google highlighted that it fine-tuned a model to understand Wendy's product names, like "JBC" for the company's junior bacon cheeseburger product. — Kif Leswing
Google touts chatbot that can help people use its cloud
Google is showing off a chatbot that can help developers — and non-developers — use its cloud-computing services.
The company is calling it Duet AI for Google Cloud. While working within Google's cloud console, users can type out questions about the available services and receive answers on the spot, so there's less of a need to look up information elsewhere online.
Google trails Amazon and Microsoft in the public cloud market, but neither of its chief rivals have released a chatbot that can do what Google says Duet AI can accomplish.
The Google PaLM 2 large language model at the core of the chatbot draws on information from Google's cloud documentation files, sample code, YouTube transcripts and podcasts, according to Richard Seroter, director of developer relations and outbound product management at Google's cloud business. Seroter spoke to reporters during a media briefing earlier this week.
During the briefing, Google aired a futuristic video showing the chatbot might one day be able to suggest that code be moved to a different cloud region to save money, set up testing infrastructure and recommend an additional cloud service to help with speed.
For developers, Duet AI for Google Cloud will be able to complete code as they type or suggest lines of code from their written comments. A similar capability is already available from the CodeWhisperer service from Amazon Web Services and from Microsoft's GitHub Copilot.
The tool can help with code reviews, Thomas Kurian, CEO of Google's cloud business, said during the keynote.
—Jordan Novet
Google is bringing generative A.I. directly to Search
Google is bringing generative AI features to search with a product called "Converse." It'll be able to synthesize search results from complex queries, although it isn't clear which users will be able to access it or when it'll roll out.
It's part of a new effort called Search Generative Experience, or SGE. It'll be available to Labs users over the "coming weeks," and users can access it by tapping on the Labs icon in the Google app or in the desktop version of Google Chrome.
Like many of the other features that Google executives have touted at I/O, there's a heavy emphasis on integrating with other Google products and with Google's search index.
The demo showcased how searching for a new bike to purchase or planning a family-friendly, dog-ready day trip could be helped with Converse.
Like some of the other features introduced in Workspace, Converse will offer a "conversational mode" where users can "ask" follow-up questions.
A big disclaimer at the top of the demo results: "Generative AI is experimental."
-- Rohan Goswami
Google says A.I. will create entire docs and spreadsheets
Google's online office suite, Workspace, will soon allow users to generate full documents and fill out spreadsheets with AI.
Google's "Help Me Write" feature that can write essays and sales pitches will be included in Google Docs for all people, the company said on Wednesday. Another example the company gave was automatically writing job descriptions for online ads by prompting a "senior level job description for a textile designer."
Google also demoed the software, based on PaLM 2, filling out an entire spreadsheet with three made-up dogs and notes for a fictional dog walking company. Google says you can take the structure and put in your own data and details.
Google Slides can generate AI images inside the app, Google said. The example on stage created images of a slice of pizza being dipped into fondue.
The new features will be available later this year for subscribers to a service called Duet AI, the company said. — Kif Leswing
Google removes waiting list for Bard chatbot
Google is upgrading its Bard artificial intelligence chatbot, making it more widely available and giving it a variety of new capabilities.
In March, Google started letting users in the U.S. and the U.K. test Bard, a rival to ChatGPT, the viral chatbot that Microsoft-backed startup OpenAI released in November. Now Google is opening access to the broader public, releasing it in over 180 countries and territories.
Last week, Microsoft scrapped the waiting list for its revamped Bing search engine, featuring a chatbot powered in part by OpenAI's GPT-4 large language model, which is at the heart of ChatGPT.
Google has moved Bard to the new PaLM 2 large language model to provide smarter answers to user questions. Models like PaLM 2 and GPT-4 are trained on reams of text data and can come up with human-like responses to questions and commands.
Bard will be able to respond to queries in English, Japanese and Korean, and Google will expand availability to the top 40 languages soon, Jack Krawczyk, Bard product lead at Google, said during a media briefing with reporters earlier this week.
Google is taking its time in opening up access to Bard because of a commitment to AI responsibility and alignment, and awareness of the limitations of the large language models that power chatbots like Bard and ChatGPT, Krawczyk said.
Bard will soon include images in responses, and in the next few months Google will make it easier to prompt the chatbot with images through the Google Lens tool, Krawczyk said. For example, a person can point a smartphone at a drawer full of arts and crafts supplies and ask what can be made with them.
Over time, Krawczyk said, Bard will be able to bring information from Google Maps, Google Docs, Google Sheets, and Gmail into conversations. Bard will also be able to show responses with help from third-party tools such as Adobe's Firefly generative AI service, he said, which can create images inspired by people's text descriptions.
Integrations with Spotify, Walmart, Redfin, Uber Eats, Tripadvisor and ZipRecruiter will appear in time as well, Google vice president Sissie Hsiao said during Wednesday's keynote.
Other new Bard features include a dark theme designed to make it easier on the eyes, and the ability for people to export responses to Gmail, Google Docs, Google's Colab interactive coding tool and third-party collaborative programming app Replit. —Jordan Novet
Protesters at Google question privacy practices around abortion location data
Protestors are flying a banner over the Shoreline Amphitheater reading "Protect Abortion Privacy," according to CNBC reporters on the ground.
"Google committed to deleting location data about health care visits including abortion care, but they are STILL retaining this sensitive data," the protestors say on their website. — Kif Leswing
Google announces its next-generation large language model, PaLM 2
Google said it has built a new general-purpose large language model called PaLM 2 that can generate a wide variety of human-like text in response to human input. CNBC reported on Monday that Google was planning to announce PaLM 2 during the I/O conference this week.
The model is skilled at math, software development, language translation reasoning and natural language generation. "It's better than our previous state-of-the-art language models," said Zoubin Ghahramani, vice president for Google DeepMind, told reporters during a media briefing earlier this week. Google announced the LaMDA model in 2021 and the more powerful PaLM model in 2022.
PaLM 2 uses Google's custom artificial intelligence chips and is more efficient to operate than previous models, Ghahramani said. The new model is able to work with over 20 programming languages, including the longstanding Fortran, and over 100 spoken-word languages, Ghahramani said.
Around 20 Google products now use PaLM 2, he said, adding that a lightweight version can run on a mobile device.
Ghahramani declined to specify the exact sources it used to train PaLM or the number of parameters the model uses. OpenAI has taken a similar approach to disclosure with its GPT-4 model. —Jordan Novet
Top Google executives spotted at the event
Many of Google's top leaders are at I/O, including the company's heads of HR and search, CNBC reporter Jennifer Elias spotted.
Immersive View will give Maps users a way to see their chosen route with A.I. driven information.
The new AI powered functionality will pull in air quality information, weather, and will visualize a chosen route for Google Maps users. CEO Sundar Pichai showcased a Manhattan bike route that visualized a bike ride alongside the West Side Highway, with AI-backed renderings of traffic, weather, and the route itself.
The demo suggested flyover functionality and that the feature would helpful information for travel superimposed over the user interface. Pichai said the functionality would roll out to over a dozen major cities in the coming months.
— Rohan Goswami
Google will use AI to help write emails inside Gmail
Google's first big feature announcement could make it much easier to write emails — or it could fill your inbox with low-quality computer generated messages.
Google will integrate generative AI into Gmail to help expand emails quickly, called "Help Me Write." Google's example was filling out a letter to get a full refund from an airline. Google has settings to make the generated text longer or shorter, and it uses context from previous messages to help write the copy.
"Just type in the prompt of what you want, hit create, and a full draft appears," Pichai said. — Kif Leswing
Google CEO Sundar Pichai takes the stage
Google's CEO has taken the stage and he's signaling that they'll be talking about artificial intelligence all day.
"As you may have heard, AI has had a very busy year," Pichai said.
"With generative AI, we're taking the next step," he added.
Expect the Google Pixel Fold, Pixel 7a and talk about new Android features
Google is also likely to discuss updates to Android, which is the most used operating system in the world with 3 billion active devices. Android 14 is expected to launch sometime this summer and the company could spend some time talking about some of the new consumer-facing features that will be available.
Google's also expected to unveil new hardware under its Pixel brand, like the Pixel Fold, Pixel 7a, and a Pixel tablet. Google's Pixel phones only have about 2% of the U.S. market, according to Counterpoint Research. -- Kif Leswing
Google opens with music partially made with A.I.
Ahead of the start of Google I/O, attendees are getting a free concert from indie electronic artist Dan Deacon.
Deacon's song, "When I'm done dying," was made partially by Google's generative AI tools, he said. The background was made by an image generator reacting to the song's lyrics.
"I'm going to play some songs, and a lot of the content is going to be made using Phenaki, Bard, and MusicLM," Deacon said. — Kif Leswing
A.I. should be center-stage at Google I/O 2023
This year, lots of Google's announcements are expected to center around artificial intelligence. AI has become the hottest part of the technology industry, partially due to attention from OpenAI and rival Microsoft, which has integrated so-called "large language models" that can spit out fluent text and image generators that can spit out original pictures into its Bing search engine.
But Google was heavily involved in the creation of these technologies, in particular, the transformers architecture that ChatGPT and other rivals use. It has its own chatbot, Bard, which is currently limited only to a select number of beta testers. It also has one of the strongest AI departments, which was recently reorganized to become Google DeepMind. -- Kif Leswing
We're here at Google I/O 2023
We're here at Google I/O 2023 seated and ready for the event to begin.
The keynote should last a couple of hours and announcements will likely run the gamut from software to hardware news. But, as we've already said, artificial intelligence is likely going to be the key focus of Google I/O 2023.
The company is planning to announce a number of generative AI updates, including launching a general-use large language model (LLM), CNBC reported on Monday. -- Jennifer Elias
