Trending Questions on AI and Emerging Technologies

Now that AI functionality has reached table stakes level, the goalposts for staying competitive have moved and multiplied. Service providers’ competitive objectives should match their strengths and business goals and may range from market leadership to technological breakthrough to relevance and a favorable position for future opportunities. Whatever the goal, pivot from functional use-case-oriented AI to delivering real business outcomes for mission-critical initiatives.

Gartner has identified four challenges all providers need to understand to stay competitive in the AI Vendor Race:

• Monitor evolving AI market opportunities.The AI race is transforming today’s breakthrough technology into tomorrow’s competitive battleground, displacing established providers and creating new revenue streams. Make it your business to track the size and trajectory of AI markets that are critical to your business.

• Find the AI value sweet spot. AI ROI will be a pipe dream until providers can close the gap between their AI investments and buyers’ needs. AI is changing technology’s basis for value away from providing tools for tasks and toward applying data-intensive solutions to realize business outcomes.
• Prepare for competitive disruptors. Small providers are flipping the competitive hierarchy by grabbing market share from the top 100 publicly held tech providers. As the rules change, so do the players. Keep your competitive edge by harnessing agentic AI and remaining relevant.
• Adapt to changing economic and business models. AI’s cost structures are changing license, maintenance and service revenue models. Transition to consumption- or outcome-based models that meet customers’ expectations of value. Adapt product pricing models to not only drive profit but maintain customers’ trust.

By 2030, future AI will see 800% growth in data center transformer workloads. A holistic innovation strategy will be critical to deliver escalating compute power while navigating acute power constraints. Product leaders should concentrate on five areas:

• The GenAI power crisis: Providers must assess the risk of power shortages and rising costs and prioritize ways to lower power requirements. Semiconductor innovation will address about 35% of this power challenge; cross-ecosystem innovation — including advanced cooling and software optimization — will be just as critical.
• Advanced efficiency: Prioritize technologies that enable higher performance, lower power consumption and greater design flexibility. Accelerate chip design automation through GenAI-assisted electronic design automation (EDA) tools.
• Networking infrastructure scale: Address GenAI’s massive data transfer demands with high-speed networking fabrics. Look to transformative technologies (e.g., interchip optical interconnects, emerging standards such as Ultra Ethernet) to tackle future GenAI workloads.
• Edge and endpoint AI: Shift toward endpoint devices and the edge to optimize energy efficiency, privacy and user experience. Prioritize energy-efficient designs that balance model size with on-device capabilities and power resources.
• Embedded AI: Expect embedded AI to improve performance and gain a competitive edge in five physical domains ─ autonomous vehicles (AVs), head-mounted displays (HMDs), smart home products, wearables and smart robots.

AI-driven, multimodal applications fueled by domain-specific language models (DSLMs) are the next wave. Investment in diverse, robust datasets from internal, external and alternative sources will feed the demand for intelligent apps with predictive accuracy.

Gartner anticipates future processes that are intelligently prompted or autonomously executed using actuated models and agents. This ability to create on-demand data, information, decisions or outputs will redefine our conception of applications.

Product leaders should explore four areas of differentiation:

• Deep enterprise AI integration: Embed AI into enterprise software, IT services and comprehensive operational workflows to ensure measurable efficiency gains and proactive decision making.
• Intelligent simulation: The shift toward more autonomous, integrated and comprehensive simulation environments will require providers to invest in intelligent simulation.
• Synthetic data sourcing: By 2030, use of synthetic data will surpass real data. Partner with leading synthetic data and simulation technology firms to boost data generation capabilities.
• Specialized AI intelligence: Enterprise investments in Earth intelligence will increase to more than 50% of total global spending by 2029. Providers should offer custom-made AI solutions for scientific and industrial breakthroughs in specialized, data-intensive domains.

Synthetic data will surpass real data as the foundation for business decision making by 2030. In the absence of large quantities of unstructured and synthetic data, product leaders will be hard pressed to develop, train, test and demo AI-powered solutions. These data types eliminate the roadblocks presented by real-world data (e.g., incompleteness, existing bias, the need for confidentiality and compliance with data protection regulations). They enable intelligent simulation that accelerate testing and deployments.

Providers that partner with synthetic data players will differentiate themselves with innovative products and services enabled by complex simulations. This will open new market opportunities and allow technology and service providers (TSPs) to deliver products that address emerging needs with precision and foresight.

Other benefits include: 

• The ability to generate high-quality, fit-for-purpose data, which will empower smaller companies to identify untapped opportunities overlooked by larger companies.
• Collaboration among synthetic data providers, simulation technology firms and data marketplaces, which will fuel the creation of integrated solutions across industries.
• Modeling hypothetical scenarios, which will allow companies to improve decision making and optimize workflows in real time, yielding cost savings and operational agility.
• Human resources that will be more strategically allocated, as reliance on real-world data diminishes and synthetic data becomes a core asset.
• Synthetic data, which will eliminate traditional data challenges, allowing for enhanced scenario planning and customer engagement.

The challenge of defending the global attack surface grid (GASG) has surpassed the abilities of the detection-and-response approach. By 2030, due to the rapid growth of the GASG, there will be more than 1 million documented cybersecurity Common Vulnerabilities and Exposures (CVEs).

Also by 2030, preemptive cybersecurity solutions will account for 50% of IT security spending, replacing traditional stand-alone detection-and-response solutions as the favored defense against cyberthreats. Providers that deliver preemptive cybersecurity capabilities — ability to deny, deceive or disrupt bad actors — will gain a competitive advantage versus those that don’t.

Providers can harden their defenses by:

• Defending the expanding GASG through developing and integrating AI-powered preemptive cybersecurity solutions. With the GASG’s continuously shifting collection of digital assets come new attack vectors and complexities that challenge cybersecurity capabilities and will surpass reactive cybersecurity measures.
• Developing preemptive cybersecurity solutions tailored to specific industries, applications or threat vectors. Success will come to those that offer deep domain expertise and strategically partner to offer comprehensive security coverage.
• Forging alliances with companies specializing in predictive analytics, DSLMs, intelligent simulation technologies and preemptive cybersecurity solutions will enable you to counter the evolving threat landscape.

Preemptive cybersecurity is no longer optional. Security product leaders must act immediately to embed it as a fundamental imperative in their roadmaps.

Keep a keen eye on potential pitfalls and possibilities to thrive during a time of disruption. Among the pitfalls that can keep a GenAI project from moving beyond the proof-of-concept stage are poor data quality, inadequate risk controls, escalating costs and unclear business value.

To get a handle on the competition:

• Future-proof your GenAI compute infrastructure in the face of grid power constraints and an uncertain chip supply chain. With new, algorithm-aligned silicon architectures disrupting the market, establish firm agreements with suppliers that have a presence in multiple regions where distributed local power is available. Consider improving the efficiency of AI software with open-source models and quantization techniques.
• Incorporate intelligent devices into AI workflows to take advantage of faster processing, increased memory and greater autonomy. Such innovations present opportunities to automate repetitive tasks and safely navigate high-risk situations.
• Be meticulous about governance and accountability requirements. With such high security stakes, due diligence and proper testing are worth the time invested. Make it your business to understand the EU AI Act, which categorizes AI applications by risk level and imposes regulations on that basis (including banning AI systems it defines as posing unacceptable risks to the public). 

In the race to embrace AI, product leaders will have to be agile, insightful and able to spot disruptive trends that represent major market opportunities. Product leaders must assess the impact of disruptive use cases, align them with long-term strategies and seize early opportunities to stay competitive. Several disruptive AI trends are already redefining the technology landscape:

• Multimodal GenAI and domain-specific language models are bridging the gap between AI and business value by enabling interactions across data modalities and offering context-specific solutions.
• Cross-modal AI systems and polyfunctional robots are integrating technology with human activity in new ways. Cross-modal vision models will surpass human-level scene understanding and predictive capabilities. Polyfunctional robots will have the flexibility to adapt to unforeseen tasks.
• Cybersecurity is shifting to preemptive cybersecurity ─ using advanced AI and machine learning to recognize and mitigate threats before they materialize.
• Disinformation security will be critical in combatting deepfakes and protecting brand integrity.
• Intelligent simulation, combined with synthetic data, is becoming foundational for cybersecurity decision making. It offers a safe, scalable and cost-effective way to test product security against real-world attacks through digital twins.
• Earth intelligence combines smart satellites, advanced image analytics and ground sensors with AI.
• Autonomous AI devices with on-device GenAI processing offer benefits like personalization, privacy and reduced latency. Sensor fusion boosts autonomous capabilities by combining data from multiple sources like synthetic sensors. Technological advancements in AI models and sensors are making self-driving vehicles feasible and creating new business models.

The days of proprietary models are numbered as open-source models and domain-specific capabilities gain ground.

• Open-source models are successfully competing with proprietary models, bringing customization within reach, fostering experimentation and giving product leaders more opportunities for innovation.
• The size and power of large language models (LLMs) is giving way to the strategic precision of small language models (SLMs), which are easier to train and perform as well or better for most enterprise tasks but at a much lower cost. SLMs may be deployed in resource-constrained environments or on-premises for better latency, security and privacy.
• Gartner predicts that DSLMs will replace LLMs for AI used in a business context. DSLMs’ proprietary data and industry-specific insights will deliver immense business value reinforced by superior accuracy, reliability and cost-efficiency.
• AI is also shifting from general pattern recognition to logical deduction and strategic planning, which is expected to advance agentic AI. The biggest obstacle to adopting agentic AI is gaining customer trust. Gartner advises product leaders to prioritize time to trust (TTT) with flexible agentic architecture that features human-in-the-loop controls and domain-specific reasoning capabilities.

IT spending for DSLM-driven software and solutions will snowball as enterprises increasingly adopt and chain multiple specialized AI models. AI marketplaces will become important as centralized platforms for purchasing specialized models and datasets, driving cost savings, efficiency and innovation.

The AI services market of 2030 will be unrecognizable to today’s providers. Gartner predicts a global surge in spending that will transform AI service solutions into self-serve platforms or software solutions, which Gartner predicts will make up 80% of the market.

In addition to investing in talent and building partnerships, service providers that want to be serious contenders in this space must fulfill three requirements:

• Revolutionize offering roadmaps and deliver AI innovation. AI agents will be central to evolving services as they will transform client operations by automating complex tasks and redesigning workflows. To boost solution sales and adoption, focus GenAI innovation on knowledge management, video analytics and advanced conversational AI.
• Align strategies to the most promising AI opportunities. Before service providers can identify the high-impact use cases that will bring quick wins, they’ll have to address the poor data quality and ineffective data strategies that are common among enterprises.
• Operationalize and monetize AI. Providers that can deliver quantifiable business outcomes (e.g., operational efficiency, cost reduction) will have the upper hand. To reach this end, they will first have to embrace the same type of reinvention they offer to clients — rethinking operations, commercial and monetization strategies and value delivery.