Google on Wednesday, February 5, released Gemini 2.0, its most capable suite of artificial intelligence models.
The mentioned virtual product from the specified technology giant is available to everyone. It’s worth mentioning that in December, the company provided access to its advanced digital brainchild powered by artificial intelligence to developers and trusted testers. The technology giant has also wrapped some features into its products.
Google’s new suite of artificial intelligence models includes 2.0 Flash. The technology giant described this virtual product as a workhorse model that is optimal for solving large-scale tasks involving a high volume and high frequency of execution.
The mentioned suite of AI models also includes 2.0 Pro. This virtual product is experimental. The technology giant has stated that 2.0 Pro is designed for coding performance.
Another digital product included in the specified suite of artificial intelligence models is 2.0 Flash-Lite. The company described this virtual brainchild as the most cost-efficient AI model to date.
For 10 cents, Gemini Flash provides developers with one million tokens for text, image, and video input. Flash Lite, a more cost-effective version of this virtual product, gives users the same features for 0.75 cents. It is worth clarifying that tokens refer to each individual unit of data processed by the artificial intelligence model.
The ongoing releases are the move within the framework of Google’s strategy aimed at making significant financial injections into machine intelligence agents. The so-called arms race in the space of the artificial intelligence industry is a process in which both technology giants operating at the global level and startups are actively involved.
Currently, the tendency is being observed, in which Meta, Microsoft, Amazon, Anthropic, and OpenAI move towards agentic machine intelligence or digital cognitive models capable of completing complex multi-step tasks on behalf of the user. In this case, an algorithm is applied that provides that artificial intelligence fully performs the task received from the consumer, without the need to clarify or describe in detail each stage of the relevant process.
In a message posted on Google’s blog in December, the technology giant noted that over the past year, it has invested in developing more agentic AI models. Also, in the relevant context, the company drew attention to the fact that such forms of existence and functioning of machine intelligence are characterized by an improved understanding of the world around the user, the ability to think through actions several steps ahead, and a function that provides for operating within specific tasks on behalf of the consumer. Moreover, a message published in December contained information that Gemini 2.0 has new advances in the area of multimodality and the use of native tools. Among other things, in this case, it implies the possibility of performing tasks such as native image and audio output. Also in December, Google noted that Gemini 2.0 allows the technology giant to develop new artificial intelligence agents that bring the company closer to its vision of a universal virtual assistant.
The main competitor of the Internet search giant in the race for elaborating machine intelligence agents is Anthropic. The mentioned startup is backed by Amazon and was founded by former OpenAI research executives. In October, Anthropic stated that its artificial intelligence agents can use computers as part of the same approach that humans take in the context of interacting with relevant devices. In this case, it implies the practice of leveraging computers to complete complex tasks. As part of using computers, the technology from Anthropic interprets what is displayed on the screen, selects buttons, navigates websites, enters text, and executes tasks through any software and real-time internet browsing. The relevant information was contained in a message published by the startup.
Jared Kaplan, Anthropic’s chief science officer, during a conversation with media representatives, emphasized that the mentioned tool can interact with computers in basically the same way that people use. Also in this context, it was separately noted that the specified virtual product can complete tasks with dozens or even hundreds of steps.
OpenAI released a similar feature called Operator. This tool allows users to automate tasks such as vacation planning, filling out forms, making restaurant reservations, and ordering groceries. The startup, backed by Microsoft, described Operator as an agent that can go to the web to perform tasks for the consumer.
OpenAI also introduced Deep Research, which allows an artificial intelligence agent to compile complex research reports and analyze questions and topics of the user’s choice. It is worth noting that Google launched a virtual product with the same name and a similar functional purpose in December. The tool from the technology giant can serve as a research assistant exploring complex topics and compiling reports on behalf of the user.
It is worth mentioning that back in December, the media reported that Google would introduce several artificial intelligence features in early 2025. At that time, the company’s chief executive officer Sundar Pichai stated at a strategy meeting that it doesn’t always have to be the first in history, but it needs to execute well and really be the best in class as a product. He also underlined that, in his opinion, that’s what 2025 is all about.
It is worth noting that against the background of the active development and scaling of artificial intelligence, the issue of cybersecurity is becoming more relevant. In the context of the relevant issue, a circumstance that generates sensitive consequences is that fraudsters also have access to AI tools. For this reason, the activities of cybercriminals have become more sophisticated. Personal awareness of users is important to counter the mentioned threat in the virtual space. For example, an Internet search query such as how to know if my camera is hacked will allow anyone to get information about signs of unauthorized access to a personal device. Digital literacy is an effective tool for countering cybercrime.
