As we approach 2025 and beyond, the TMT industry is poised for significant advancements, driven by rapid generative AI adoption. However, several critical gaps must be addressed to unlock its full potential. These include balancing investments in Gen AI infrastructure with monetization, addressing gender disparities in Gen AI usage, managing the energy consumption of Gen AI data centers, and tackling trust concerns around deepfake content. Additionally, discovering optimal uses for Gen AI in media and gaming, along with harnessing the power of Gen AI agents to operate in real-time, remains vital. Further challenges persist in streaming video and cloud spending, while new opportunities arise with innovations like Gen AI chips in smartphones and PCs, enhanced sports infrastructure, and telecom consolidation.
Eight Gaps Highlighting 2025 as a “Gap Year” for TMT
- The Gen AI Infrastructure and Monetization Gap Massive investments are being made in generative AI infrastructure, with tens of billions spent on chips and hundreds of billions on building data centers for model training and inference. While some companies offering Gen AI enterprise software are seeing revenue growth, the returns are currently dwarfed by the investments—often by a factor of 10 or more. Although many argue that underinvestment poses a greater risk than overspending, the gap between costs and returns continues to widen.
- The Gen AI Data Center Electricity and Sustainability Gap Generative AI data centers demand unprecedented amounts of power, ideally from low-carbon sources. This has created a mismatch between energy requirements, grid capacities, and corporate sustainability goals. Despite efforts by hyperscalers, chip manufacturers, and utility providers to bridge this gap, it is expected to persist into 2025.
- The Gen AI Gender Gap Women are currently less likely than men to utilize generative AI tools for work or leisure. A significant factor is a lack of trust, although women’s adoption of Gen AI is expected to catch up to men’s in certain markets by the end of the year.
- The Gen AI Deepfake Trust Gap The rise of deepfake content—spanning images, video, and audio—is eroding public trust in digital media. To address this, the generative AI ecosystem must implement robust measures, such as labeling AI-generated content comprehensively and developing reliable, real-time detection systems for fake media. The cost of creating convincing deepfakes is dropping, and detection systems must become equally cost-effective to bridge this gap.
- The Studio Gen AI Usage Gap There is a notable gap between the expectation that major studios will widely adopt Gen AI for content production and the current reality. While studios are exploring these tools to reduce costs, save time, and expand reach, concerns around intellectual property rights related to generative content have led to cautious adoption.
- The Autonomous Gen AI Agent Gap Autonomous AI agents capable of reliably completing tasks and managing workflows are a promising frontier. Pilot programs for these agents are set to launch in 2024, but widespread adoption remains uncertain for 2025.
- The Streaming Video Gap Media companies anticipated that consumers would maintain multiple streaming subscriptions indefinitely. Instead, many households are reducing the number of services they subscribe to, opting for bundled offerings and dropping less-used subscriptions. This shift has forced streaming platforms to focus on partnerships and aggregators to close the gap in revenue growth.
- The Cloud Spending Gap Cloud adoption was initially marketed as a cost-saving measure, but poorly managed and decentralized spending has led to rising expenses. To address this, organizations are turning to FinOps (Financial Operations) to align spending with cost-saving goals, potentially saving billions and narrowing the gap between expectations and reality.
2025 Key Topics: A Quick Overview
Women and Generative AI: Adoption Rising, Trust Lagging
While women’s use of generative AI is expected to match or surpass men’s in the U.S. by 2025, a significant trust gap persists.- Trends: Deloitte predicts women’s adoption of generative AI will double by 2025, reaching parity with men in many regions.
- Challenges: Women report lower trust in AI providers regarding data security, which may limit engagement.
- Solutions: Tech companies should prioritize increasing trust by improving data security, reducing bias, and fostering diverse workforces.
Powering AI: Sustainable Energy for Data Centers
Generative AI’s increasing power demands are pushing data centers toward greener energy solutions.- Growth: AI-driven infrastructure is predicted to account for 2% of global electricity consumption by 2025.
- Challenges: The demand for reliable, clean energy faces regulatory and grid limitations.
- Recommendations: Optimize AI chips, adopt carbon-free energy sources, and collaborate across industries for sustainable solutions.
Transforming Stadiums into Economic Hubs
Sports stadiums are evolving into community-centered destinations, driving socioeconomic growth.- Investment Trends: Nearly half of new stadium developments by 2025 will be in North America and Europe.
- Outcomes: These projects foster public-private partnerships, enhance community engagement, and diversify revenue streams.
Autonomous Generative AI Agents: The Future of Workflows
Generative AI agents capable of autonomous operations are emerging as productivity tools.- Adoption: By 2025, 25% of AI-using companies will pilot autonomous AI agents, rising to 50% by 2027.
- Applications: These agents will revolutionize areas like customer support, software development, and cybersecurity.
Deepfake Disruption: A Growing Cybersecurity Challenge
The rise of deepfake content demands robust detection and verification systems.- Threats: The authenticity of online media is at risk as deepfake technologies improve.
- Approach: Industries are leveraging cryptographic metadata and AI tools to counter fake content, similar to cybersecurity efforts.
Optimizing Cloud Spending with FinOps
Enterprise cloud spending is set to exceed $825 billion globally by 2025.- Strategies: Companies adopting FinOps can reduce waste, optimize spending, and save $21 billion globally.
- Cultural Shift: Linking cloud expenses directly to business outcomes ensures better financial accountability and decision-making.
On-Device AI for Smartphones: A Game Changer?
Generative AI could redefine smartphones, making them smarter and more efficient.- Market Growth: Global smartphone shipments are expected to grow by 7% in 2025.
- Opportunities: New AI capabilities like intelligent assistants could redefine user experiences if effectively implemented.
Generative AI in Hollywood: Slow but Strategic Adoption
Major studios are cautious about using generative AI in creative workflows.- Current Use: Less than 3% of production budgets will go toward generative AI in 2025.
- Focus Areas: Studios are leveraging AI for operational efficiencies, such as marketing, localization, and talent management.
Video Streaming: A Shift Toward Aggregators
Standalone streaming services are losing momentum as aggregators gain traction.- Consumer Trends: The average number of standalone services is expected to decline post-2024.
- Impact: Aggregated services can reduce costs and create a more sustainable ecosystem.
Wireless Telecom Consolidation: A Gradual Shift
In fragmented markets, mergers among smaller wireless operators are gaining support.- Trends: Europe and Asia are seeing regulatory approval for telecom consolidation, enabling sustainable growth.
- Outcome: Consolidation is expected to create stronger networks with better long-term viability.
Emerging Topics to Watch
- Generative AI and Cybersecurity: A double-edged sword, generative AI presents both risks and opportunities in cybersecurity.
- Chiplets Technology: Advanced semiconductor architectures like chipsets are driving innovation in AI and high-performance computing.
- Silicon Photonics: Enabling high-speed data transfer, silicon photonics is set to revolutionize AI data center communications.