Google may be back at the frontier of artificial intelligence, but investors and consumers are increasingly focusing on what people can build with the technology as personal agents surge in popularity.
The company unveiled its long-awaited Gemini 4 Argon model on Wednesday, promising major advances in coding, cybersecurity, and complex tasks. According to industry benchmarks, Argon ties OpenAI on a key cybersecurity test and posts leading results in software engineering.
Argon's introductory pricing of $2 per million input tokens and $10 per million output tokens also matches OpenAI's newly discounted GPT-6.1 Sol model. A token is equivalent to about three-quarters of one word.
While Google is stepping up its game in the model battle, rival Meta is racking up millions of downloads for its Muse app, which launched last month and soared to the top of Apple's App Store, topping ChatGPT. Earlier this week, OpenAI rolled out Dots, its personal agent offering, capitalizing on demand from businesses and consumers for tools that can carry out tasks — like scouring emails, booking travel and organizing expenses — on a user's behalf.
Meanwhile, the attention at the frontier model level has turned to safety, with Anthropic CEO Dario Amodei sparking a firestorm three weeks ago by urging the top AI labs to slow the pace of development as concerns intensify about their potential dangers.
Over the past three months, Alphabet's stock is down about 6%, while Meta is up 19% over that stretch, sparked by September's rally, which was the sharpest for any month since 2022.
Analysts at JPMorgan Chase wrote in a note on Thursday that Google needs a significant advance in its personal agent offerings to generate consumer enthusiasm. Bank of America identified the potential for Gemini 4 to strengthen Google's cloud business and existing products, while providing a foundation for a future personal agent.
In the consumer market, Google has a massive home-court advantage.
Billions of people use Google products, including Gmail, Calendar, Chrome, and Search. Their messages and schedules are already stored within Google's ecosystem, potentially giving the company a substantial edge in developing an assistant capable of working across those services.
Yet Google's personal agent, Spark, remains limited to paying subscribers, while Muse is free with usage caps. As of Sept. 30, Muse had reached more than 5 million downloads, according to Sensor Tower.
In a background briefing with CNBC, members of Google's agent team outlined Spark's existing capabilities and the company's approach to expanding access, and acknowledged that Muse's free availability has helped consumers discover and experiment with it.
Launched in May at Google's I/O developers conference, Spark can work across Gmail and Calendar, navigate websites through Chrome, and complete tasks such as filling out online forms. It's also available through Google's mobile and desktop apps.
But unlike Muse, Spark can't make outbound phone calls or complete purchases on a user's behalf. Instead, it takes users through the purchasing process before handing control back to them for final approval.
Can Argon help Spark?
Meta's approach has also introduced complications.
Reuters reported in September that the company had experimented with a human concierge system in which contractors handled some phone calls when Muse couldn't complete them autonomously. Internal privacy concerns prompted Meta to suspend the experiment, while its automated phone-calling feature remains in beta.
The episode has raised privacy concerns associated with granting AI assistants access to so much sensitive personal information.
A Meta spokesperson said in an email that internal testing "is core to the product development process" and is being done to get feedback on safety and privacy protections needed for a broader release.
"We're working with merchants to continue improving this potential calling feature, and will only roll it out when it's ready and with the proper disclosures," the spokesperson said.
Google is now evaluating whether its latest frontier model could strengthen Spark.
In an interview with CNBC, Tulsee Doshi, head of product for Gemini, highlighted Argon's ability to tackle complicated assignments that require multiple steps and can run for extended periods. She said Google is still evaluating where Argon can be deployed most effectively.
Personal agents don't need the most powerful frontier models to complete everyday tasks. And prior to Argon, Google spent much of the summer releasing cheaper, lighter models designed to handle routine requests at scale. Doshi said those models remain essential for everyday agents, while more powerful frontier models can provide the additional reasoning capabilities required for complex assignments.
There's also the money. Running powerful AI models is expensive, particularly for agents that operate continuously and perform multiple tasks. Google's introductory Argon pricing is competitive, but the company said its published rates will eventually double to $4 per million input tokens and $20 per million output tokens.
One challenge for Google is determining how to combine its models into an agent that can handle increasingly sophisticated tasks without making the service prohibitively expensive to operate.
Argon also isn't broadly available yet.
Google is participating in a voluntary U.S. government safety-testing process and initially restricting access to select cybersecurity defenders and enterprise cloud customers. The company plans to expand availability to developers, enterprise customers and consumers, but hasn't announced a public release date.
That's limiting the independent testing of Argon's capabilities. Bloomberg reported Wednesday that some Google employees had questioned the model's real-world coding performance despite strong benchmark results.
Google disputes that characterization, and told CNBC that employees across the company have been testing versions of Gemini 4 for weeks, with some receiving unlimited access.
But the greater urgency is on the agent side, as other companies race to bring AI to the masses.
As JPMorgan analysts wrote, "Google has struggled to translate ongoing product velocity into splashy releases that resonate with consumers, capture sustained mindshare, & meaningfully drive sentiment around AI leadership."
—CNBC's Jonathan Vanian contributed to this report.
WATCH: Google announces new frontier model called Gemini 4 Argon