At HubSpot for Startups' annual AiSummit in San Francisco, industry leaders gathered for a provocative discussion about how AI agents are poised to fundamentally transform the SaaS ecosystem. The session brought together Amjad Masad from Replit, Mark Roberge from Stage 2 Capital, Jia Li from LiveX AI, and Nikunj Handa from OpenAI, moderated by Adam Biddlecombe from Mindstream.
The New DNA of AI-Native SaaS Companies
Velocity is Everything
According to Masad, the defining characteristic of AI-native SaaS companies is velocity. "Companies need to be exponentially faster than they used to be," he emphasized. This means rapidly moving from first product to second to third product, because traditional moats are disappearing.
Key principles for AI-native companies:
- Stay lean: Question every headcount addition and consider AI tools before hiring more people
- Self-disrupt: Move fast enough that you disrupt yourself before competitors do
- Embrace efficiency: Your competitor will be using AI tools, so you must too
From Chatbots to Action-Taking Agents
Li drew a compelling analogy to explain the evolution from basic AI tools to agents: "AI agent is like riding a self-driving car, while chatbot is more like a map navigator. AI agent will really take action when the green light is on and help you to move forward."
This shift enables productivity improvements across all company functions:
- Product and design teams
- Marketing and engineering
- Sales and customer support
Every role can leverage AI agents to become more productive and move faster.
The Incumbent's Dilemma
Why Established Companies Will Struggle
Roberge painted a stark picture for incumbent SaaS companies, drawing parallels to historical disruptions. "Ninety percent of the CRM market in 1995 was owned by Siebel," he noted, highlighting how difficult it is to defend against massive tech shifts.
The challenge isn't just about redefining products—it's about redefining organizations in an AI-first way. Companies need to rethink:
- Internal processes and workflows
- Organizational structure and roles
- The fundamental nature of the C-suite
The Future Leadership Team
In an agentic world, traditional functional boundaries disappear. Future executives will likely resemble today's operations professionals more than traditional managers:
- Future CROs may look more like today's revenue operations people
- Future CPOs may resemble SVPs of DevOps
- Cross-functional expertise becomes essential as agents handle frontline work
The Evolution of Human-AI Interaction
From Reactive to Proactive AI
Handa outlined the progression of AI interaction models:
- Current state: Type a message, get a response
- Background tasks: Kick off deep research, get notified when complete
- Proactive AI: AI figures out what you need when you wake up, reacts to emails and Slack messages, and solves problems before you know they exist
Who Wins in the AI-First World
The panelists agreed that success will depend on:
- Learning agility: Ability to adapt quickly to new tools and products
- Good judgment: Knowing what's valuable and what isn't
- Generative thinking: Becoming an "idea generator" who can try multiple approaches simultaneously
- Interdisciplinary expertise: Breaking down traditional functional silos
As Masad put it: "Everyone becomes an entrepreneur. You wake up in the morning and your goal is to really have impact."
Rethinking Traditional Business Models
The Death of Seat-Based Pricing
Traditional SaaS metrics are being rewritten:
- Enterprise sales reps producing $10 million instead of $1 million annually
- Gross margins of 95% instead of 80%
- Rule of 100 instead of Rule of 40
Pricing model evolution:
- Outcome-based pricing: Charge $1 per closed ticket rather than per seat
- Usage-based pricing: Align costs with value creation
- Continuous pricing optimization: Treat pricing as an ongoing project, not a one-time decision
The Venture Capital Disruption
With AI enabling smaller teams to create more value, traditional VC models face disruption. Masad questioned: "I don't even know why you would go to a VC anymore" for many SaaS companies, given how quickly companies can now monetize and grow revenue.
Practical Strategies for Entrepreneurs
Building at the Edge of What's Possible
Rather than focusing on traditional moats, Masad advocated for continuous innovation: "The ultimate moat is being a year or two ahead every time."
His advice for entrepreneurs:
- Start building on emerging technologies like computer use models now
- Develop expertise at the frontier of what's possible
- Be ready when the next model breakthrough makes your startup suddenly viable
Rethinking Customer Support and Workflows
The panel highlighted how AI enables complete workflow reimagination. Instead of managing support tickets reactively, AI can:
- Anticipate customer issues based on behavior patterns
- Provide proactive assistance before problems occur
- Unify presale, support, and onboarding functions
- Deliver personalized rescue experiences for at-risk customers
Traditional Advice That's Now Wrong
The panelists identified outdated business wisdom:
- Years of experience matter most: Fresh thinking often trumps traditional experience
- Focus on brand name investors: Traction is the only thing that matters now
- Stick to established SaaS metrics: All the numbers are changing with AI
The Path Forward
The session concluded with a clear message: rewrite everything. From work schedules to growth expectations to organizational structures, AI enables us to question fundamental assumptions about how businesses operate.
For entrepreneurs, the opportunity is massive—but so is the need for speed. As Li noted: "Company needs to be moving fast, and AI agent is the one that every individual can manage in the future."
The SaaS companies that will win aren't just those that add AI features—they're the ones that completely reimagine what's possible when humans and AI agents work together.
~~~~~
AI Disclaimer: AI helped us summarize the key points from these videos, and our editorial team reviewed everything to make sure it's clear and correct.
