The Future of Team Topologies: When AI Agents Dominate
Can human and AI team members connect?
I've been contemplating the future of technology a lot recently. The speed at which artificial intelligence (AI) has evolved in the last two years is nothing short of phenomenal. What the data-centre laden ChatGPT3 did just two years ago, is now surpassed by models like Qwen2.5 and Phi4 running on nothing more than a powerful, but consumer laptop. My MacBook Pro M3 Max will easily run a model that is 30% smarter, and 100% faster than GPT3.
As AI keeps advancing rapidly, we're on the brink of a big change where AI agents could make up 50%, 70%, or even 90% of our teams. This brings up important questions about the future of work, how we collaborate, and how organisations are designed. How will ideas like Conway's Law, Dunbar's Number, and context switching work in a world where AI is everywhere?
The Rise of AI-Dominated Teams
Picture teams where AI agents do most of the work—designing, writing, testing, and debugging code all by themselves. These AI agents can even update themselves to get better and more efficient. This isn't just a fast change; it's a revolutionary one. It makes us rethink how teams work and what roles humans will play in a future driven by AI.
Conway's Law Reimagined
Conway's Law says that the way systems are designed reflects how organisations communicate. In a world full of AI agents, this law changes. AI systems can be set up to communicate perfectly without the social limits humans have. But the challenge is making sure these AI systems still meet human goals and values. As AI becomes more independent, organisations need to design their structures carefully to make sure AI helps rather than hinders human collaboration.
Dunbar's Number: An Obsolete Constraint?
Dunbar's Number suggests there's a limit to how many stable social relationships a person can maintain—around 150. In AI-dominated teams, this limit doesn't matter. AI agents don't need social bonds or emotional connections; they work purely on logic and efficiency. This could allow teams to grow much larger than before, but it also raises questions about empathy, trust, and human connection in a workplace filled with AI.
Context Switching: An AI Advantage
Humans often struggle with context switching—the mental effort needed to switch between tasks. AI agents, however, are great at multitasking and can change focus quickly without any mental strain. This could lead to big productivity gains, as AI can move between tasks smoothly, optimising workflows in ways humans can't. But this also means organisations need to rethink how they structure work to make the most of AI's strengths while keeping human workers engaged and productive.
The Human-AI Symbiosis
In a future where AI agents dominate teams, the role of human workers will change. Humans will need to focus on tasks that require creativity, empathy, and ethical judgment—areas where AI still falls short. This partnership could lead to more innovative and efficient teams, but it needs careful management to ensure human skills are valued and not made obsolete.
Successful Human-AI Partnerships
Even with AI taking on a bigger role, successful collaborations between humans and AI show a promising future. For example, in healthcare, AI helps doctors by analysing large amounts of data to find patterns and suggest diagnoses, while doctors provide the necessary human touch for patient care and ethical decisions. In creative fields, AI can generate ideas or draft content, which humans then refine and improve with their unique perspectives.
These partnerships show that when used effectively, AI can enhance human abilities rather than replace them. By focusing on what each does best—AI in data processing and pattern recognition, humans in creativity and empathy—we can create teams that are not only more efficient but also richer in innovation and insight.
The Social Aspect of AI-Dominated Teams
Even as AI takes on a larger role, the social aspects of team dynamics remain crucial. Here’s how they manifest:
Human-AI Interaction: Humans must collaborate with AI to set goals, interpret results, and make ethical decisions, ensuring technology aligns with organisational values.
Team Culture and Values: Human oversight is necessary to instill culture and values within AI systems, guiding their behavior and decision-making processes.
Communication and Coordination: Effective communication protocols between humans and AI are essential for seamless interaction and feedback loops.
Trust and Reliability: Building trust in AI requires transparency and accountability, with human mechanisms for monitoring and auditing AI behavior.
Ethical Considerations: Humans must manage the ethical implications of AI decision-making, addressing biases, ensuring privacy, and maintaining compliance with legal standards.
Learning and Adaptation: Human team members play a key role in refining AI algorithms through continuous learning and feedback.
Empathy and Emotional Intelligence: While AI excels at data processing, humans bring empathy and emotional intelligence to the table, essential for understanding customer needs and fostering a positive work environment.
Provocative Questions for the Future
What happens when AI agents become the primary decision-makers in teams? How do we ensure these decisions align with human values and ethics?
Can organisations maintain a sense of culture and community when AI dominates team compositions? What role will human leaders play in fostering an inclusive and cohesive work environment?
How do we balance efficiency gains from AI with the need for human creativity and innovation? Will AI-driven teams stifle or enhance human potential?
Conclusion: A Brave New World
The future of Team Topologies, dominated by AI agents, presents both opportunities and challenges. As AI continues to evolve, organisations must navigate this new landscape thoughtfully, ensuring that technology enhances rather than diminishes the human experience at work. By reimagining principles like Conway's Law, Dunbar's Number, and context switching in an AI-driven world, we can create teams that are not only more efficient but also more aligned with our deepest values and aspirations.
Connect with Andy Spamer here.
About the author:
Andy Spamer is an organizational agility coach and delivery lead based in Western Australia. He has worked with many of the largest blue-chip companies in Australia, and has had exposure to global delivery projects in mining and finance. Andy is a Team Topologies Advocate and also a proud geek who is always tinkering with tech and software.
Connect with Andy here.
