The dust reaches $ 6 million.

soilThe two-year-old artificial intelligence platforms that help institutions to build artificial intelligence agents who are able to complete the entire workflow tasks, from annual revenues of $ 6 million-an increase of six times more than one million dollars just one year ago. The rapid growth of the company indicates a shift in the adoption of the AI ​​from the Foundation from the simple Chatbots towards advanced systems that can take concrete measures through business applications.

San Francisco’s startup company announced on Thursday that it was chosen as part of the “Claude” ecosystem from the Anthropor, highlighting a new category of artificial intelligence companies that builds specialized institutional tools higher border models instead of developing its artificial intelligence systems from the zero point.

“Users want more than just conversation interfaces,” Gabriel Hubert, CEO and co -founder of DUST said in an interview with Venturebeat. “Instead of creating a draft, they want to create the actual document automatically. Instead of obtaining meetings, they need to update CRM records without manual intervention.”

The DUST platform exceeds the Chatbot style that has controlled the adoption of the early institution. Instead of just answering questions, artificial intelligence agents in DUST can create GitHub problems automatically, scheduling calendar meetings, updating customer records, and even paid symbol reviews based on internal coding standards-while maintaining safety protocols at the institution level.

How artificial intelligence agents convert calls to Github tickets and CRM updates

The company’s approach is clear through a concrete example, describing Hubert: a sales company from business to business using multiple dust agents to process sales call texts. One of the agents analyzes the sales arguments that resonated with the prospects and update of the battle cards automatically in Salesforce. At the same time, another agent determines customer features, and he appoints them to the product road map, and in some cases, he automatically creates GitHub tickets for small features that are ready for development.

“Each call version will be analyzed by many agents,” explained. Hopart. “You will have an improved agent for the sales battle cards that will be seen in the arguments made by sales, which were strong and seemed to resonate with the possibility, and this will go and feed on an operation on the Salesforce side.”

This level of automation is enabled by Form Context Protocol (MCP)A new standard developed by anthropologist that allows artificial intelligence systems to communicate safely with external data sources and applications. Guillaume Princen, President of Europe, the Middle East and Africa in Man, described MCP as “like a USB-C connector among artificial intelligence models and applications”, enabling agents to access company data while maintaining safety limits.

Why Claude and MCP operate the next wave of AI’s automation to the institution

Dust’s success reflects wider changes in how institutions approach artificial intelligence. Instead of building dedicated models, companies such as DUST take advantage of the basic models that are increasingly capable – especially the Claude 4 suite into humans – and integrate them with specialized synchronization programs.

“We just want to give our customers access to the best models,” Hubert said. “I now think that Antarbur is early in the foreground, especially on models related to coding.” The company receives customers from 40 to 50 dollars per month per month and serves thousands of work spaces ranging from small startups to large companies that contain thousands of employees.

Claud models in humans have seen a particularly strong adoption of coding tasks, as the company has reported 300 % growth in the use of Claude icon during the past four weeks after the latest CLAUDE 4 models. “OPUS 4 is the most powerful coding model in the world,” Princesen pointed out. “We were already driving the coding race. We are strengthening it.”

The security of the institution becomes complicated when artificial intelligence agents can

The shift towards artificial intelligence agents who can take real measures through business systems provide new safety complications that were not present with simple Chatbot applications. DUST is dealt with this through what Hubert calls the “original permission layer” that separates the rights to access data from the rights to use the agent.

“Extrinary creation, as well as data and tools is part of the movement of navigation to alleviate sensitive data when artificial intelligence agents work through multiple business systems,” explains the company in technical documents. This becomes very important when agents have the ability to create GitHub problems, update CRM records, or amend documents via the Foundation Technology Pack.

The company implements the infrastructure of institutional classification with the policies of retaining data in the anthropoor, which ensures that sensitive business information that is processed by artificial intelligence agents is not stored by the model provider. This addresses a major concern for institutions that are considering adopting artificial intelligence on a large scale.

Increase the construction of original startups from artificial intelligence on basic models instead of their own creation

DUST growth is part of what Antarbur calls the ecological ecosystems for “original startups of artificial intelligence” – which cannot exist mainly without the advanced AI capabilities. These companies do not build business not by developing their artificial intelligence models, but by creating advanced applications at the top of the current foundation models.

“These companies have a very strong feeling of what these final customers need for the specific use state,” Brenson explained. “We offer their tools to build and adapt their products to these specific clients and use the situations they are looking for.”

This approach is a major shift in the structure of the artificial intelligence industry. Instead of each company you need to develop its own AI capabilities, specialized platforms such as DUST can provide a synchronization layer that makes AI models strong useful for specific business applications.

What is the revenue growth signs of $ 6 million on the future of the institution’s programs

The success of companies such as DUST indicates that the AI ​​Enterprise market exceeds the experimental phase towards practical implementation. Instead of replacing human workers in bulk, these systems are designed to eliminate routine tasks and switch the context between applications, allowing employees to focus on higher value activities.

“By providing the beginnings of Amnesty International that makes all the company’s workforce more intelligent in addition to the correct permission system, we are setting the foundations for the operating system of the agent that is resistant in the future.”

The company’s customer base includes the institutions convinced that artificial intelligence will mainly change the commercial processes. “The common thread among all customers is that they are emanating from the future and convincing that this technology will change a lot of things.”

When artificial intelligence models become more capable and protocols like MCP mature, the distinction between artificial intelligence tools that simply provide information and those who take action may become a major discrimination in the institution market. DUST rapid revenue growth indicates that companies are ready to pay distinctive prices for artificial intelligence systems that can complete real work rather than helping this.

The effects of individual companies extend to the broader structure of the institution’s programs. If artificial intelligence agents can integrate the workflow with automation and automation through separate business applications, it can reshape how institutions think about program purchases and the workflow design – which reduces the complexity that has been complex for a long time.

Perhaps the most amazing sign of this shift is how to naturally describe Hubert as artificial intelligence agents not as tools, but as digital employees who appear to work every day. In the business world that has spent decades in linking systems with applications platforms and integration platforms, companies such as DUST prove that the future may not require delivery of everything – just teaching artificial intelligence to move in the chaos that we have already created.

Daily visions about business use cases with VB daily

If you want to persuade your boss at work, you have covered VB Daily. We give you a scoop from the inside about what companies do with obstetric artificial intelligence, from organizational transformations to practical publishing operations, so that you can share visions of the maximum return on investment.

Read with us privacy policy

Thanks for subscribing. Check more VB news bulletins here.

An error occurred.