In partnership with

Aloha in 2025,

I’ve disconnected for almost two weeks, going into the middle of nowhere with my high school crew. 19 adults, 5 kids and 1 dog - daily sauna sessions, tennis, jogging, cooking, and never ending conversations until late hours… Pure bliss!!! I already miss that and can’t wait for another reunion! 😍 

Batteries recharged! Ready for 2025 🚀 Are you? Because it’s gonna be wild…

Let's dive into something transformative happening in our AI world.

I’ve been consuming a lot of AI-related content during the break and want to share with you a curated list of things worth watching/listening/reading. (list at the end of the post)

Enjoy 🔥 

Automate your meeting notes

• Condense 1-hour meetings into one-page recaps

• Say goodbye to post-meeting data entry

• Claim 90 days of unlimited AI notes today

Get started free

Will you be attending LEAP conference in 🇸🇦 ?

Last year we were exhibiting; this year, I have the honor join as a speaker! Super stoked about this, as LEAP is the largest tech event in the region, and one of the biggest tech events in the world.

If you are coming, DM on LinkedIn me and let’s catch up 🙂 

The future belongs to AI Agents. Does it?

The hype around AI agents is hitting fever pitch. NVIDIA's CES showcase, OpenAI's breakthrough o3 model, and Google's agent architecture whitepaper all point to 2025 as the year agents move from experiments to enterprise deployment. But what's actually changing, and are we ready?

Three major developments from last weeks are reshaping what's possible:

• NVIDIA's agent ecosystem provides the first complete enterprise platform. Their Blueprints enable custom agents that can analyze data, understand video, and coordinate through a unified orchestration layer. Early adopters like SAP and ServiceNow signal serious enterprise interest.

• OpenAI's o3 model achieved 88% on the notoriously difficult ARC-AGI benchmark - matching human performance. While expensive to run today, rapidly falling compute costs suggest capable reasoning agents are closer than we thought.

• Google's Gemini 2.0 represents a fundamental shift - moving from models that simply understand information to ones that can act on it. Their Project Mariner achieves 83.5% success on real-world web tasks, while Project Astra brings agent capabilities to mobile and AR devices.

Meanwhile, startups are racing to build specialized agents for everything from coding to customer service. A wave of YC companies promise "AI employees" for various business functions.

🟡 Several key challenges remain

But we've seen promising tech face adoption headwinds before. Here is what’s different now - pros and cons regarding rapid AI Agents adoption.

Historical Precedent

Recent research from PwC indicates that AI is compressing the productivity J-curve, allowing for a quicker uptick in productivity compared to previous technologies like computers and electricity. In sectors with high AI exposure, productivity growth has been nearly five times greater than in those with lower exposure. This suggests that AI may be driving a productivity revolution much faster than the historical trends observed with past technologies, where significant time was required for productivity gains to materialize. (source)

Limited Enterprise Readiness

As of 2024, approximately 72% of organizations report incorporating AI into at least one business function, a significant increase from just 55% in the previous year. Moreover, recent Accenture research shows that the number of companies with fully modernized, AI-led processes has nearly doubled from 9% in 2023 to 16% in 2024. This reflects a growing readiness among enterprises to adopt AI technologies compared to the earlier statistic of only 1.5% of businesses investing in AI. (source)

Infrastructure Costs

The costs associated with implementing AI projects remain substantial. McKinsey estimates that customizing an existing AI model can cost around $10 million, while developing a model from scratch may reach up to $200 million. Despite these high costs, many enterprises are beginning to see significant returns on their investments, with reports indicating that organizations achieving AI-led processes experience 2.4 times greater productivity and 2.5 times higher revenue growth compared to their peers. (source)

Integration Complexity

Integration challenges persist as many organizations still struggle to transition from pilot projects to full-scale implementation. Recent data shows that over half (54%) of AI projects face difficulties in achieving operational scale. However, as organizations become more familiar with AI technologies, the number of companies reporting successful integrations is increasing, suggesting gradual improvements in handling integration complexity. (source)

Skills Gap

The skills gap remains a critical barrier to AI adoption, with 75% of enterprises citing a lack of AI/ML skills as their primary challenge. However, job growth requiring these specialist skills has outpaced overall job growth by a factor of 3.5 since 2016. This indicates a growing recognition of the need for skilled professionals in the AI domain, suggesting that while the gap exists, there is also an increasing focus on developing these essential capabilities within the workforce. (source)

😄 The Path Forward

2025 looks different from previous AI hype cycles. The major platforms are building complementary pieces of the agent ecosystem:

• Google focuses on consumer-facing agents and web automation

• NVIDIA builds enterprise infrastructure and development tools

• OpenAI pushes the boundaries of agent reasoning capabilities

Their enterprise partnerships provide clear deployment paths. More importantly, they're not just launching technology - they're building infrastructure for practical business adoption.

The winners will be companies that balance ambition with execution:

• Start small with focused agent use cases

• Build internal capabilities gradually

• Use established platforms to accelerate development

• Maintain control over agent deployment and oversight

The future belongs to AI agents - but that future arrives through steady progress, not overnight revolution. 2025's platforms and tools finally make that progress possible. The question isn't if agents transform business, but how quickly companies can overcome adoption challenges to reap the benefits.

👇️ Listen/Watch/Read about Agents

That’s all for today!

Until next time,

Iwo