AI is paving the way for a new type of organization – a Frontier Firm

As artificial intelligence (AI) evolves from a tool to a true assistant, its role in the workplace expands, fundamentally transforming how enterprises operate. Microsoft’s latest research identifies a new type of organization known as the Frontier Firm, where on-demand intelligence requirements are managed by hybrid teams of AI agents and humans.

Also: Will AI replace software engineers? It depends on who you ask

On Wednesday, Microsoft published its 2025 Work Trend Index Annual Report, which combines survey data from 31,000 workers across 31 countries, Microsoft 365 productivity signals, LinkedIn hiring and labor trends, and expert insights to give employees and business leaders a comprehensive view of the work landscape. 

The focus of this year’s report is the concept of a Frontier Firm, which helps researchers explore what work structure will be like in the AI era. As expected, the differences from today are vast. Yet, the report found that every organization will be on its journey to become a Frontier Firm in the next two to five years.

Also: The best AI for coding in 2025 (and what not to use)

“We went through this exercise as a team, where we were like, if we were starting today, this was the first day in the history of work, there was no history preceding it, how would we design work?” said Alexia Cambon, Future of Work Researcher at Microsoft. “We were surprised by how many assumptions we had about how work should be done.”

The journey

The report identifies three phases for an organization’s transformation into a Frontier Firm. The first phase involves employees interacting with assistants to work more efficiently. In Phase 2, humans use AI agents, which function as more of a “digital colleague,” carrying out tasks under human direction and freeing up their time. 

<!–>

–>

“The phase most of us are in right now is AI is showing up as an assistant at work,” said Cambron. “But we definitely see the journey will get towards the Frontier Firm, where agents are a part of the workforce and are helping you do net new things, net new types of knowledge work, and those are two very different types of organizations.”

Also: Why OpenAI’s new AI agent tools could change how you code

The final phase (Phase 3) involves humans working with a team of AI agents, which can run entire business processes and workflows. According to the report, this evolution of AI into knowledge work will mimic how AI-enabled software development has progressed from general assistance in coding to chat interfaces and onto agents that can carry out the task on the user’s behalf. 

<!–>

Productivity gains

The report identified real productivity gains from implementing AI into organizations, with one of the biggest being filling the capacity gap – as many as 80% of the global workforce, both employees and leaders, report having too much work to do, but not enough time or energy to do it. 

Now, organizations can “buy intelligence on tap,” which is made possible through AI agents that act as digital labor and help companies scale as needed, according to the report. 

The report found that 46% of leaders say their companies use agents to automate workflows or processes. The exact details of agent-human interaction will look different per function, with some tasks handled independently by agents and others requiring more human involvement.

However, adding AI isn’t enough, as the speed businesses evolve is outpacing the rate at which people work, with a larger shift needed to rethink knowledge-based tasks. 

Also: AI agents aren’t just assistants: How they’re changing the future of work today

According to the report, business leaders need to separate knowledge workers from knowledge work, acknowledging that humans who can complete higher-level tasks, such as creativity and judgment, should not be stuck answering emails. Rather, in the same way working professionals say they send emails or create pivot tables, soon they will be able to say they create and manage agents – and Frontier Firms are showing the potential possibilities of this approach. 

The report found that only 844 employees out of the 31,000-person sample worked at companies that met the five traits that comprise Frontier Firms: organization-wide AI deployment, advanced AI maturity, current agent use, projected agent use, and a belief that agents are key to realizing return on investment on AI. 

Out of these firms, 71% of workers reported that their company is thriving, compared to just 37% globally; 55% say they’re able to take on more work compared to 20% globally; and 90% report they can do more meaningful work compared to 73% globally. These workers are also less likely, 21% compared to 38% globally, to fear AI will take their job. 

Workforce changes 

Beyond the positive outcomes from AI, there will be inevitable changes in the workforce’s appearance and operation. Every industry will experience the transformation differently, with AI set to generate new jobs and replace others. 

Also: As AI agents multiply, IT becomes the new HR department

Of the business leaders surveyed, 45% said expanding team capacity with digital labor is a top priority in the next 12 to 18 months. The top three areas for accelerated AI investment include customer service, marketing, and product development. 

Even after expanding their digital workforce, the researchers found that a key component enterprises need to consider to maximize benefits includes getting the human-agent ratio right. This concept refers to striking a balance between having too few agents per person and underutilizing their potential, or overwhelming the human capacity with too many agents per person. 

As an example of balance, the report cites a Harvard study that found that an individual with AI outperforms a team without the technology, but a team with AI outperforms everyone. Cambron said this trend highlighted how an organization will get better outcomes when it has a bigger team paired with AI, rather than using the technology to reduce the need for humans altogether. 

“I don’t think we need to reduce headcount and replace it with AI. I think we need to maintain headcount and augment it with AI to get the best outcomes,” said Cambron. 

Also: Will synthetic data derail generative AI’s momentum or be the breakthrough we need?

Another element of workplace change is seeing AI as a teammate rather than a tool, allowing the technology to take on higher-level tasks such as managing projects. However, 52% of respondents currently see AI as a command-based tool, and 46% see it as a thought partner.

“We need to work towards having a more uniform understanding of how to use it, because if we’re doing it like a search engine or a spreadsheet, it’s not going to deliver the same return as if we start using it more as a digital colleague, where we’re iterating and we’re brainstorming together,” said Cambron.

Maximizing AI agents’ performance involves learning about management concerns, including delegating, iterating, prompting, and refining the technology. The need for someone to manage these teams of AI agents has led to the evolution of a new role, the agent boss, responsible for the best performance. 

“Everyone will need to manage agents,” said Cambron. “I think it’s exciting to me to think that, you know, with agents, every early-career person will be able to experience management from day one, from their first job.”

Also: Employers want workers with AI skills, but what exactly does that mean?

As a result in this shift of duties, the traditional organizational chart may also see some changes, being replaced by a Work Chart, which Microsoft describes as an “outcome-driven model where teams form around goals, not functions, powered by agents that expand employee scope and enable faster, more impactful ways of working.” 

Microsoft compares this system of working to movie production sets in which tailored teams get together to perform the project and disband once the work is completed. In the same way, different AI agents will work together to meet a goal quickly and then shift onto a new project without having to reorganize actual employees. 

Humans with AI skills are also in demand, with 47% of business leaders listing upskilling their existing workforce as a top priority in the next 12 to 18 months, 51% of managers saying AI training or upskilling will become a key responsibility for their teams within five years, and 35% considering hiring AI trainers to guide employee adoption in the next 12 to 18 months. 

Although 33% of business leaders surveyed are considering a headcount reduction, Microsoft predicts new roles will evolve. Most business leaders, 78% for non-Frontier Firms and 95% for Frontier Firms, are considering hiring people for AI-specific roles, such as AI agent specialists and AI strategists in marketing, finance, customer support, and consulting, to prepare for the future. 

“We really see this new type of intelligence as additional, not substitutions,” said Cambron. “I believe some jobs will go away, it’s a natural order of events, but we absolutely believe we’re not going to run out of work – I think new jobs will be created, and current jobs will evolve.”