OpenAI takes the functionality of its chatbot ChatGPT to a new level.
The mentioned digital product is the world’s most popular chatbot powered by artificial intelligence. In fact, it was the debut of ChatGPT in November 2022 that marked the beginning of the so-called AI boom, which is currently a global process and the main technological tendency of the present historical moment.
The expansion of the list of features of the specified digital product, which became known on Thursday, January 23, is associated with the addition of a function designed to automate tasks such as vacation planning, filling out forms, ordering groceries, and making restaurant reservations. The new virtual instrument is called Operator. OpenAI describes this functional solution as an agent that can go to the web to perform tasks for consumers. The startup also noted that Operator was trained to interact with buttons, menus, and text fields that people use daily. In this case, it means the process of interaction carried out on the Internet.
Operator can also ask additional questions. The corresponding action is intended to further personalize the tasks performed, including, for example, information required to log in to other websites. Users can take control of the screen at any time.
In a message that was published on the OpenAI blog on Thursday, it was noted that Operator is one of the first agents that is a functional form of artificial intelligence capable of doing the work for the client independently. It was noted that in this case, the person gives the task, and the digital instrument executes it.
Currently, Operator is only available to ChatGPT Pro users in the United States. Access to the new virtual instrument can be obtained on the online page Operator.ChatGPT.com. OpenAI announced that it intends to eventually provide access to the agent to Plus, Team, and Enterprise users. The startup also plans to integrate Operator into ChatGPT.
Moreover, OpenAI reported that the new digital instrument is currently facing problems when performing certain tasks, including calendar management and slideshow creation.
The startup stated that users can opt out of collecting some training data by turning off the setting to Improve the Model for Everyone in ChatGPT. This means that personal information in Operator will not be applied to train its models.
The startup also announced that users can delete all browsing data and log out of all sites with one click in the privacy section.
Operator competes directly with release from Anthropic, an Amazon-backed artificial intelligence startup that developed a chatbot called Claude. In October, the mentioned startup presented Computer Use. As part of this functional artificial intelligence solution, AI agents can use computers to complete complex tasks. The mentioned agents interact with computers as part of an approach to performing appropriate actions, similar to the algorithm applied by humans. In this case, it is possible to perform a variety of tasks. Anthropic stated that artificial intelligence agents can interpret information on a computer screen, select buttons, enter text, and navigate websites. Also, in this case, it is possible to perform tasks through any software and Internet browsing in real-time.
Jared Kaplan, Anthropic’s chief science officer, stated during a conversation with media representatives that the mentioned instrument can use computers in almost the same way as humans do that. It was also noted that in this case, it is possible to perform tasks with dozens or even hundreds of steps.
It is worth noting that the generative artificial intelligence market is currently on an active scaling trajectory. The largest players in the relevant functional space, in addition to OpenAI and Anthropic, are such technology giants as Microsoft, Meta, Google, and Amazon. The revenue of the generative artificial intelligence market is expected to exceed $1 trillion within a decade.
Moreover, AI as a technology continues to develop intensively. Artificial intelligence has already demonstrated impressive abilities related to processing huge amounts of information, including through an analytical approach, and generating original content based on user prompts. AI is also already being used as a highly accurate and highly efficient working tool in many industries. Moreover, some futurological forecasts suggest that artificial intelligence in the course of its evolution may surpass the human mind in terms of cognitive abilities. Against the background of the existing opportunities and prospects of AI, global interest and active investment in the area of digital cognition are logical.
Google has agreed to make additional financial injections into Anthropic in the amount of more than $1 billion. This month, the media released information according to which the mentioned startup is currently in the late stages of talks about raising $2 billion in a funding round at a $60 billion valuation led by Lightspeed Venture Partners.
OpenAI is currently pushing toward developing the future of artificial general intelligence. It is worth clarifying that AGI is a configuration of digital thinking that is equal to or even superior to humans in terms of cognitive abilities. In this case, superiority in the context of mental activity can relate to a specific list of tasks, or it can be absolute. It is worth noting that AGI still continues to be what can be called an indefinite higher goal. It is currently unknown what limits of cognition of the surrounding world, including in the cosmological aspect, will have an artificial intelligence configuration that surpasses the human mind. At the same time, there is no guarantee that people will be able to understand the corresponding cognitive limits, which are beyond the capabilities of their thinking.
Scale AI chief executive officer Alexander Wang, during a conversation with media representatives, stated that he defines AGI as powerful artificial intelligence systems that can use a computer in the same way as a human. In his opinion, it will likely take from two to four years to reach this level of technology development.
It is worth noting that against the background of the active development of artificial intelligence, the issue of cybersecurity is growing in urgency. In this context, it is important to pay attention to the fact that fraudsters also have access to machine intelligence tools. These criminals use artificial intelligence as part of their activities. The result of this practice is that detecting fraudulent activity in the virtual space has become more difficult. To counter this threat in the cyber environment, personal awareness of users is important. 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 the device.
The main purpose of criminals in the virtual space is to steal money through online banking apps. In some cases, cybercriminals target the personal data of their victims for subsequent compromise. At the same time, the typical ultimate goal of these actions is to steal money.
Artificial intelligence can also be used by cybercriminals to generate fake media content. Sometimes such videos, which involve supposedly real people, force the victims of the mentioned criminals to perform certain actions. For example, a boss may instruct a subordinate to transfer money to a specific bank account. The victim, unaware of the trick, carries out the assignment. Then it turns out the non-fake boss did not say anything like that and had no corresponding intentions. Using images or videos of real people, cybercriminals act on their behalf in the virtual space. It is worth noting that this problem is not what can be called a hopelessness factor for potential victims. Startups are developing artificial intelligence-powered tools to combat deepfakes and other forms of virtual crime related to the use of AI. Any technology can be used both in constructive scenarios and to achieve destructive goals. The application of artificial intelligence by cybercriminals is not evidence of the harmfulness of AI. Machine intelligence has many useful scenarios that significantly exceed fraudulent harm that can be overcome. Moreover, AI was not originally designed for deception and virtual theft. Recently, there has been a tendency to adopt artificial intelligence in the area of cybersecurity.
