An in-depth analysis of the latest AI trend report from the “Queen of the Internet”: where are enterprises and individuals heading?

In today's world of rapid technological advancement, a report has once again ignited our imagination of the future - the AI ​​Trends Report by "Internet Queen" Mary Meeker. This report not only reveals the amazing progress of AI technology, but also depicts its future development blueprint from a forward-looking perspective. For each of us and every company, this is not only a detailed analysis, but also a strategic guide for future survival and development.

We have analyzed the report in depth and explored how to apply these insights in real business environments. We hope that this article can reveal the key signals under the wave of AI.

Unprecedented scale and speed: AI is spreading like wildfire

"Change is happening at an unprecedented speed" - this sentence at the beginning of the report accurately summarizes the core characteristics of the current development of AI. This is not an exaggeration, but is supported by a series of amazing data:

• The “rocket speed” of user growth
  Take ChatGPT, for example. Just 17 months after its public release, this phenomenal app is expected to reach a staggering 800 million weekly active users by April 2025. This growth trajectory is almost unparalleled in the history of technological development and fully demonstrates the enthusiastic pursuit of AI technology by consumers and business users.

• Big Tech Gambles
  The capital expenditure (Capex) of the Big Six in the United States in 2024 will reach $212 billion, a year-on-year increase of 63%. The report directly pointed out that the core driving force behind this huge investment is AI. They are not just "testing the waters", but "betting" on the future.

• The "blowout" of the developer ecosystem
  Take chip giant NVIDIA as an example. The number of developers in its ecosystem has increased sixfold in just seven years to 6 million, which marks the rapid popularization and application of AI tools and frameworks among the developer community.

The report even gives a formula: AI user growth + usage growth + capital expenditure growth = unprecedented.  What does this mean for the vast majority of companies? It means "urgent action"! The market will not wait, and we must seize opportunities as quickly as possible instead of responding passively.

The “compound interest” effect of technology: the core engine driving change

The amazing development of AI is not the result of a single technological breakthrough, but the result of the “compounding effect” of technological progress in many aspects:

• Hash rate surge
  Training computing power is growing at a rate of 360% per year.

• Algorithm optimization
  Contributes approximately 200% performance improvement every year.
• Supercomputing Evolution
  Supercomputer performance increases by 150% every year.
• Model iteration acceleration
  Over the past four years, the rate at which powerful new AI models are released has increased by an average of 167% per year.

This “compounding” growth places extremely high demands on infrastructure:

• Cloud services are advancing rapidly
  The annual revenue growth rate of global hyperscale cloud services (such as Google Cloud, AWS, and Microsoft Azure) is as high as 37%. Companies such as NVIDIA account for about a quarter of the global data center capital expenditure revenue.
• Data center construction "great leap forward"
  Over the past two years, the value of data center construction in the United States has increased by an average of 49% per year. The report mentioned an amazing case: a 750,000 square foot (about 70,000 square meters) facility was built in just 122 days.

However, a key bottleneck has also emerged: energy . The report clearly states: "There is no AI without energy." Although the energy efficiency of AI models and hardware (such as NVIDIA GPUs) has increased by about 100,000 times in the past decade, according to the "Jevons' Paradox", the sharp decline in computing costs has led to an explosive growth in usage, which in turn has caused total energy demand to continue to rise. Since 2017, global data center electricity consumption has increased by about 12% per year. The progress of AI is beginning to be constrained by power grids and energy demand.

What this means for businesses is that even when relying on cloud services, they must factor underlying cost, availability, and energy into their long-term strategies.

AI reshapes the working model: a “real and rapid” evolution

AI’s change in the way we work has gone from theory to reality, with the core being the leap in “cognitive automation” capabilities. The report estimates that in just three years, the reasoning capabilities of AI models may have jumped from the level of high school students to that of college graduates or even senior engineers.

• Enterprises pay close attention
  More than half of S&P 500 companies mentioned AI in their earnings calls. Corporate AI goals are more focused on revenue growth, building customer-facing AI applications, and developing new revenue streams rather than simply cost reduction.
• Significant productivity improvement
  The report cited cases where customer service staff using AI assistants saw a 14% increase in productivity.
• Big changes in the job market
  Over the past seven years, the number of job postings in the United States that explicitly mention AI has surged 448%, while non-AI IT jobs have fallen 9%.

• “AI First” Becomes a Strategy
  An internal memo from the CEO of language learning platform Duolingo stated bluntly: "Duolingo will become an AI-first company." This includes incorporating the use of AI into recruitment and performance evaluations, linking employee growth to automation efficiency, and even in some cases accepting small sacrifices in quality for speed and scale.

As a leader of NVIDIA said: " You won't lose your job because of AI, but you will lose your job because of those who know how to use AI. " The strategic focus of enterprises and individuals should shift to empowerment so that everyone can become "the person who knows how to use AI."

Intense competition: opportunities and challenges in the changing situation

While it is still expensive to train cutting-edge large models, the cost of inference (the cost of running a trained model) is falling rapidly . This makes powerful AI capabilities increasingly affordable, benefiting all users and developers. However, this also intensifies competition:

• Traditional AI profit model faces three threats
  Increasingly fierce competition + the rise of open source models + the rapid development of Chinese AI.
• The power of open source
  Open source models such as Meta's Llama have been downloaded billions of times, and their performance is rapidly catching up with closed-source proprietary models. They are highly favored by developers for their low cost and high flexibility.
• Technology giants fully penetrate
  Giants such as Meta (3.4 billion users), Apple (2.35 billion devices), and Google (nearly 5 billion search users) are deeply embedding AI into their existing product lines (such as Microsoft Copilot, with annualized revenue of $13 billion).
• AI application layer startups are growing rapidly
  Anysphere AI grew from $1 million to $300 million in ARR in 25 months; Harvey in the legal tech space grew from $10 million to $70 million in ARR in 15 months. The median time for AI companies to reach $5 million in ARR is 35% faster than for typical SaaS companies.
• The US-China AI “Space Race”
  The report describes the competition between China and the United States in the field of AI as a "space race" with significant geopolitical implications. China has made rapid progress in model development (such as Alibaba's Qwen and Baidu's Ernie) and has shown an advantage in inference costs.

For enterprises, this means more choices, but also means that competition is everywhere. It is critical to choose the technology stack wisely and balance capabilities, costs, and tool sources.

The future is here: the fusion of intelligent agents, multimodality and the physical world

The report further looks forward to the next stage of AI development:

• The evolution of AI agents
  From "chat response" to "active task execution". AI agents are intelligent processes that can coordinate actions across applications and services. In the past 16 months, the search volume for "AI agent" on Google has increased by more than 1,000%.
• The explosion of multimodal AI
  The number of AI models that can understand and generate text, images, audio, video and other information has increased by 1,150% in the past two years. This will give rise to a large number of practical scenarios, such as diagnosing machine failures through images, or automatically generating medical reports by combining X-rays and medical records.
• AI is accelerating into the physical world
• Autonomous driving
  Waymo has a significant share of the San Francisco ride-hailing market, and Tesla's fully autonomous driving mileage is growing exponentially (about 100 times).
• national defense
  Companies like Anduril have seen their annual revenue double.
• agriculture
  AI-driven weeding robots, etc.
• The report succinctly summarizes: AI is transforming capital assets into software endpoints .
• Connecting 2.6 billion “offline population”
  With the popularity of low-cost satellite Internet (such as Starlink, which has more than 5 million users), the 2.6 billion people in the world who are not yet connected to the Internet may soon be connected to the Internet. Their first Internet experience may no longer be a traditional browser or APP, but an AI-driven, native-language-supported multimodal intelligent interface.

These cutting-edge trends remind us that we must go beyond the current text-based AI optimization and think about how AI can deeply interact with the physical world in the form of intelligent entities and in a multi-sensory (multi-modal) manner, and design new digital experiences for "AI natives."

Our Action Guide: Embracing an AI-First Strategic Transformation

This report sends a clear signal to all businesses and individuals: AI is not another option on the technology roadmap, but the next fundamental platform shift .

To stay invincible in this transformation, or even lead the trend, “ AI first ” must become the core strategy:

1. Decisive investment
   Emulate the resolve of the tech giants and make strategic investments in AI capabilities.
2. Choose wisely
   Deeply understand the trade-offs and opportunities between cutting-edge closed-source models and fast-iterating open-source models, and make the choice that best suits your business.
3. Deep Fusion
   Embed AI into core workflows to drive productivity improvements and business growth.
4. Forward-looking layout
   Prepare for the future of intelligent agents and multimodal interactions.
5. Cultural Reshaping
   This is not just a technological upgrade, but also a comprehensive innovation of the operating model, talent recruitment and assessment standards (giving priority to talents who can collaborate efficiently with AI), team structure and even corporate culture. The case of Duolingo is worth pondering.

Conclusion: Time waits for no one, the future is already at our feet

"When AI is as deeply integrated into daily life and work as the Internet is today, where will your business department, your team, and your personal career be in a year or two?" This is not only a question for yourself, but also a soul-searching question for every company. The report clearly points out that in this rapidly changing environment, stagnation is equal to active failure. The pace of change will not slow down, but will only get faster and faster.

Now is the time to think seriously and take immediate action to become the one who makes good use of AI and build an AI-first future.