As leaders, we face critical decisions that will determine our companies' futures in the AI era. Daily, we evaluate where to invest, which partnerships to forge, and how to position our organizations for sustainable growth. Having led AI initiatives from both Silicon Valley and industrial sectors, I've witnessed firsthand how this technology is reshaping competitive dynamics across industries. Let me share insights to help you harness AI's transformative power while avoiding its pitfalls.

The Current State of the AI Landscape

The AI landscape is rapidly evolving. Google's Brain team restructuring has created internal turbulence, yet this is the same organization that gave us Transformers — the foundation of today's AI breakthroughs. Despite their challenges, Google remains a vital player in AI innovation.

Microsoft is making bold moves through its OpenAI partnership, aggressively leveraging this alliance to capture enterprise market share. Whether this IT dominance will extend to Operational Technology remains questionable, as industrial leaders resist additional "Microsoft tax." Beyond Silicon Valley, nations are pursuing AI independence — Japan has explicitly declared AI sovereignty a national security priority, investing billions in projects like Fugaku. This global diversification creates two distinct development paths: tech giants investing in broad foundation models, while enterprises focus on domain-specific AI models.

Understanding AI Trends

Foundation Models & AGI

Foundation models are becoming increasingly multimodal, integrating language, images, and video comprehension. Our perception of AGI is shifting — no longer benchmarked solely against human intelligence. Like calculators surpassing humans in numerical calculations, AGI is developing its unique capabilities independent of human cognitive patterns.

Evolution of AI Hardware

Having witnessed AI hardware evolve from CPU to GPU dominance, I see our industry approaching another transition toward Intelligence Processing Units (IPUs). These units, combining compute and memory like biological brains, align with AI's computational needs. While still emerging, expect significant adoption within 5–10 years.

Evolution of AI Software

As the AI landscape evolved, another significant shift has occurred — the movement of foundation models toward open-source. In the past, these models were proprietary, developed and closely guarded by tech giants. This created a significant barrier for companies wishing to leverage AI. Now there's a transformative trend underway, with foundation models increasingly becoming open-source, democratizing access and giving companies the ability to create custom models based on proven architectures. One notable example is Meta's Llama model: launched as open-source, it has become the progenitor of a rich variety of AI models — a "Llama family tree." Such open-source models enable companies to bypass the substantial costs of developing AI from scratch, focusing resources instead on customization and application-specific fine-tuning.

Key Business Implications and Actions

Two Strategic Paths

Businesses face a clear bifurcation in AI strategy:

  • Foundation Model Development: Large tech companies with vast resources are competing to build broad, multimodal foundation models.
  • Domain-Specific Model Creation: Most enterprises will leverage their specialized knowledge to create targeted AI models for specific tasks.

Small Specialist Models (SSMs)

Small Specialist Models — efficient, domain-specific AI solutions — represent a significant opportunity. Aitomatic's OpenSSM project is accelerating their development and adoption globally. For your business, this presents dual opportunities: creating SSMs leveraging your domain expertise, and integrating these models into your operations.

The Infrastructure Opportunity

"Picks-and-shovels" companies providing essential AI implementation infrastructure are seeing explosive growth. A compelling example is MosaicML's acquisition by Databricks. Specializing in making large-model training more efficient and reliable, MosaicML saw a surge in demand following ChatGPT's release, rapidly growing revenues and culminating in their acquisition — a testament to the value of specialized AI infrastructure providers. While many focus on end applications, some of the most promising opportunities lie in building the foundational technologies that enable others to implement AI effectively.

Actionable Advice for CEOs Like You

Strategic Positioning

  • Acknowledge AI's central role — AI isn't merely another technology initiative; it's reshaping competitive dynamics across every industry.
  • Navigate the geopolitical landscape — Regional AI investments are creating arbitrage opportunities. Position your organization to benefit through strategic partnerships and infrastructure planning.
  • Watch industry evolution — Major players like Google and Microsoft are redefining the AI landscape. Their moves signal directional shifts that will affect your competitive positioning.

Execution Framework

  • Leverage domain expertise — Your organization's unique knowledge is your greatest AI asset. While tech giants compete on general capabilities, your competitive advantage lies in domain-specific applications. Invest in Small Specialist Models tailored to your industry's particular challenges.
  • Progress from Q&A to problem-solving AI — Begin with knowledge-indexing AI that makes your corporate information accessible and searchable. This delivers immediate productivity gains — but your strategic focus should be on problem-solving AI that can evaluate options, make recommendations, and optimize processes.
  • Bridge IT and OT — IT-focused AI improves knowledge-worker productivity, but the greater ROI often lies in OT applications: AI that optimizes manufacturing processes, predictive maintenance, and physical operations. Start with IT wins but plan for OT transformation.
  • Embrace open-source — The rise of open-source foundation models like Meta's Llama creates a new economic reality. Rather than building AI from scratch, leverage these pre-trained models and focus resources on fine-tuning — reducing development costs by orders of magnitude while accelerating time-to-value.
  • Anticipate cost reduction — Just as with previous technology waves, AI infrastructure costs are declining rapidly while capabilities increase. Focus your investment decisions on applications that create lasting value rather than being deterred by today's implementation costs.

Future-Proofing

  • Prepare for hardware evolution — Plan for the transition from GPUs to IPUs. Companies pioneering this shift will gain significant competitive advantages.
  • Adopt efficient training methods — As model complexity increases, efficient training technologies become critical competitive differentiators.
  • Position for System-2 AI — Current AI lacks higher-level abstraction for planning and reasoning. Prepare for System-2 AI capabilities that will enable truly autonomous intelligence.
  • Transform, don't just adopt — Use AI as a catalyst to fundamentally rethink business processes rather than simply automating existing workflows.

Looking to the Future

The AI revolution isn't merely another technology cycle — it's fundamentally reshaping competitive dynamics across industries. The winners will be organizations that not only adopt AI but strategically integrate it to transform their operations and customer experiences. Our responsibility as leaders extends beyond implementation: we must position our organizations to leverage AI's transformative potential while navigating its challenges. The future belongs to those who act boldly today, with clarity of vision and confidence in execution. Let's lead our organizations with that mindset.