Technological Disruption Risks

Technological Disruption Risks in the context of AI search refer to the strategic threats posed by rapid advancements in artificial intelligence technologies that fundamentally challenge established market leaders through generative AI tools that reshape information retrieval patterns and user behaviors ¹. The primary purpose of understanding these risks within competitive intelligence (CI) and market positioning frameworks is to enable organizations to monitor and anticipate how AI-native entrants like Perplexity and OpenAI systematically erode the dominance of incumbents such as Google, which currently commands approximately 90% of traditional search queries ¹. This matters profoundly because it drives strategic repositioning imperatives: firms must proactively integrate AI capabilities to avoid revenue erosion from diversified search habits, with projections showing that even dominant players like Google face pressure despite anticipated ad revenue growth at 8% CAGR, underscoring the critical need for sophisticated competitive intelligence to safeguard market position ¹.

Overview

The emergence of technological disruption risks in AI search represents a fundamental shift in how organizations approach competitive intelligence and strategic positioning. Historically, the search industry operated under relatively stable competitive dynamics, with Google establishing near-monopolistic dominance through superior algorithmic retrieval and network effects built over two decades ¹. However, the advent of generative AI technologies beginning in 2022-2023 introduced a new paradigm where AI-native platforms could synthesize information rather than merely retrieve it, fundamentally altering user expectations and behaviors ¹².

The fundamental challenge these disruption risks address is the potential obsolescence of traditional search business models built on link-based navigation and click-through advertising revenue. As AI systems provide direct, synthesized answers to user queries, they reduce clicks by 20-30%, creating what industry analysts call "zero-click searches" that compress publisher revenues and threaten the foundational economics of ad-dependent search platforms ². This phenomenon forces competitive intelligence practitioners to expand their monitoring beyond traditional market share metrics to encompass user behavior shifts, technological capability assessments, and ecosystem interdependencies across retail, social, and AI-native platforms ¹².

The practice has evolved rapidly from initial dismissal of AI chatbots as novelties to recognition of existential threats requiring strategic pivots. By 2024-2025, incumbents like Google launched defensive innovations such as Search Generative Experience (SGE) to bridge retrieval and synthesis capabilities, while new entrants leveraged open-source models and conversational interfaces to capture market segments ¹. This evolution reflects a broader pattern described in Clayton Christensen's disruptive innovation theory, where initially underperforming technologies rapidly improve and displace established players through superior accessibility and efficiency ³.

Key Concepts

AI-Native Search

AI-native search refers to information retrieval systems that prioritize synthesis and contextual understanding over traditional link-based navigation, using large language models to generate direct answers rather than lists of web pages ¹². This represents a fundamental architectural shift from keyword matching to semantic comprehension and natural language generation.

Example: When a user asks Perplexity "What are the best practices for managing technological disruption risks?", the platform synthesizes information from multiple sources into a coherent narrative with inline citations, eliminating the need to visit individual websites. In contrast, traditional Google search would return a list of ten blue links requiring users to click through and synthesize information themselves. This efficiency gain—reducing user effort by approximately 40% according to user experience studies—drives adoption particularly among younger demographics who prefer conversational interfaces ²³.

Zero-Click Search Phenomenon

Zero-click searches occur when users obtain complete answers directly from search results pages without clicking through to any destination websites, fundamentally disrupting the click-based advertising model that underpins traditional search economics ². This phenomenon intensifies as AI systems become more capable of comprehensive synthesis.

Example: A marketing professional searching for "current AI search market share statistics" receives a complete answer with specific percentages and trends directly in ChatGPT's response window, sourced from multiple recent reports. The professional never visits the original publisher websites, meaning those publishers receive no traffic, no ad impressions, and no revenue despite their content being used. Industry data shows this pattern now affects 20-30% of searches, representing billions of dollars in potential revenue displacement annually ².

Competitive Moat Erosion

Competitive moat erosion describes the systematic weakening of defensive advantages that previously protected incumbent market leaders from competitive threats, particularly as AI democratization lowers entry barriers and enables rapid capability replication ¹³. In search, traditional moats included proprietary algorithms, massive index infrastructure, and network effects from query data.

Example: Google's historical advantage required decades of crawling infrastructure and billions in capital investment to build comprehensive web indexes. However, when OpenAI launched ChatGPT with browsing capabilities in 2023, it achieved comparable search functionality within months by combining GPT-4's language understanding with real-time web access through partnerships, bypassing the need for proprietary indexing infrastructure. This was further accelerated when Apple integrated ChatGPT into iOS, instantly providing the AI assistant with distribution to hundreds of millions of users—a moat-crossing event that would have been impossible in the pre-AI era ¹.

Generative AI Democratization

Generative AI democratization refers to the widespread accessibility of advanced AI capabilities through open-source models, API services, and low-code platforms that enable diverse market entrants to deploy sophisticated search alternatives without massive capital investment ¹³. This fundamentally alters competitive dynamics by enabling rapid experimentation and niche specialization.

Example: A mid-sized e-commerce company leverages open-source models like Llama 2 combined with retrieval-augmented generation (RAG) frameworks to build a specialized product search assistant for their vertical market. Using cloud infrastructure and pre-trained models, they deploy a conversational search experience in three months with a team of five engineers—a capability that would have required years and millions in investment previously. This pattern repeats across retail (Amazon), social (TikTok), and specialized domains, fragmenting search traffic across multiple platforms ¹.

Query Evolution and Conversational Interfaces

Query evolution describes the shift from keyword-based search strings to natural language questions and multi-turn conversations, fundamentally changing how users interact with information retrieval systems and creating advantages for AI-native platforms optimized for dialogue ²³. This behavioral shift particularly affects younger demographics and voice-based interactions.

Example: A Gen Z professional researching competitive intelligence methodologies engages in a 15-minute conversation with Claude, asking follow-up questions like "How does that apply to AI search specifically?" and "What tools would you recommend for monitoring these trends?" This conversational pattern—now representing 40% of daily searches among U.S. users according to voice search adoption data—creates context and personalization impossible in traditional keyword search. The user develops platform loyalty to Claude based on conversational quality, eroding Google's habitual usage patterns built over decades ²³.

Revenue Model Transformation

Revenue model transformation encompasses the strategic shift from advertising-based monetization dependent on click-through rates to alternative models including subscriptions, AI premium tiers, and direct answer monetization as zero-click searches undermine traditional economics ¹². This forces fundamental business model reassessment across the search ecosystem.

Example: Facing declining click-through rates as SGE provides direct answers, Google experiments with "AI Premium" subscription tiers offering enhanced synthesis capabilities, ad-free experiences, and priority access to advanced models—similar to ChatGPT Plus. Simultaneously, they develop sponsored synthesis where brands pay to have their products mentioned in AI-generated answers, creating a new ad format that works within zero-click paradigms. Early pilots show 8% CAGR potential but require complete sales force retraining and advertiser education, illustrating the operational complexity of revenue model pivots ¹.

Ecosystem Interdependencies

Ecosystem interdependencies describe the complex relationships between search platforms, content publishers, advertisers, device manufacturers, and regulatory bodies that amplify disruption impacts through cascading effects across the value chain ¹⁴. Understanding these interdependencies is critical for comprehensive competitive intelligence.

Example: When OpenAI partners with Apple to integrate ChatGPT into iOS, this creates ripple effects: publishers see traffic decline from Safari searches, Google faces pressure on its lucrative default search payments to Apple (estimated at $18-20 billion annually), Android manufacturers consider competitive AI integrations, and regulators scrutinize the partnership under antitrust frameworks. Meanwhile, advertisers shift budgets toward platforms maintaining click-through rates, creating a feedback loop that accelerates incumbent vulnerability. Competitive intelligence teams must map these interdependencies to anticipate second and third-order effects of technological shifts ¹.

Applications in Competitive Intelligence and Market Positioning

Strategic Horizon Scanning for Emerging AI Capabilities

Organizations apply technological disruption risk frameworks to systematically monitor AI advancement signals including open-source model releases, academic breakthroughs, startup funding patterns, and user adoption metrics ³. This enables early detection of capability shifts that could alter competitive dynamics before they manifest in market share changes.

A Fortune 500 technology company establishes a dedicated AI horizon scanning team that monitors arXiv preprints, GitHub repository activity for search-related projects, venture capital investments in AI search startups, and user sentiment analysis across social platforms. When they detect Perplexity's traffic growing 300% quarter-over-quarter in early 2024 among technical users—a leading indicator demographic—they accelerate their own conversational search development timeline by six months and adjust competitive positioning messaging to emphasize synthesis capabilities. This proactive response, informed by systematic CI, prevents market share erosion in their core enterprise search segment ¹³.

Competitive Benchmarking and Capability Gap Analysis

Firms employ structured frameworks to assess their search capabilities against both traditional competitors and AI-native entrants across dimensions including answer accuracy, response latency, conversational coherence, and source attribution quality ¹. This quantitative benchmarking informs investment prioritization and positioning strategy.

A market research firm conducts monthly blind testing where analysts pose identical complex queries to Google, ChatGPT, Perplexity, Claude, and their internal search tools, scoring responses on accuracy, completeness, and usability. Results show their platform lags AI-native competitors by 40% on synthesis quality but leads on domain-specific accuracy. This intelligence drives a positioning pivot emphasizing specialized expertise over general search, while simultaneously informing a technology roadmap to integrate retrieval-augmented generation for improved synthesis. The benchmarking also reveals that Perplexity's citation quality exceeds ChatGPT's, informing partnership discussions with the former ¹.

Scenario Modeling for Market Evolution Pathways

Organizations develop multiple future scenarios ranging from incumbent adaptation to complete market restructuring, using these models to stress-test strategies and identify robust positioning approaches that succeed across diverse outcomes ¹³. This application proves particularly valuable given the uncertainty inherent in AI trajectory predictions.

A digital advertising agency builds three detailed scenarios for 2027: "Hybrid Dominance" where Google successfully integrates SGE and maintains 70% share, "Fragmented Landscape" where traffic splits across five major platforms, and "AI Extinction" where zero-click searches eliminate 60% of publisher traffic. For each scenario, they model revenue impacts, required capability investments, and optimal client positioning strategies. This analysis reveals that diversifying client presence across multiple AI platforms and developing direct monetization capabilities provides resilience across all scenarios, leading to a strategic recommendation that clients reduce Google dependency from 80% to 50% of search marketing budgets over 24 months ¹².

Regulatory Intelligence Integration

Sophisticated competitive intelligence operations integrate technological disruption monitoring with regulatory trend analysis, recognizing that antitrust actions, AI safety regulations, and data privacy frameworks significantly influence competitive dynamics in AI search ¹. This holistic approach captures interdependencies between technology and policy.

A European search engine company tracks both Google's SGE development and parallel EU Digital Markets Act enforcement actions that could restrict Google's ability to preference its AI features in search results. Their CI team identifies a strategic window where regulatory constraints on Google combined with their own AI integration could enable market share gains in privacy-conscious segments. They position their AI search as "regulation-compliant synthesis" and capture 3% market share among enterprise customers prioritizing GDPR alignment—a niche made viable by the intersection of technological and regulatory disruption ¹.

Best Practices

Establish Continuous AI Capability Monitoring Systems

Organizations should implement automated monitoring systems that track AI search competitor capabilities, user adoption patterns, and technological breakthroughs on at least a weekly cadence, rather than relying on quarterly competitive reviews that prove too slow for AI's rapid evolution ³. The rationale is that AI capabilities can double within months, making traditional review cycles obsolete and creating strategic blind spots.

Implementation Example: A competitive intelligence team deploys a monitoring dashboard that aggregates data from multiple sources: API calls to competitor AI search platforms testing response quality on standardized queries, web scraping of user reviews and social media sentiment, RSS feeds from AI research publications, and traffic analytics from SimilarWeb showing competitor growth patterns. The system generates automated alerts when competitors show capability improvements exceeding 20% or traffic growth exceeding 50% month-over-month. This enabled one organization to detect Perplexity's emerging strength in technical queries three months before it became widely recognized, allowing preemptive positioning adjustments ¹³.

Conduct Cross-Functional AI Disruption War Games

Organizations should facilitate quarterly war gaming exercises bringing together technology, strategy, marketing, and executive teams to simulate competitive scenarios where AI-native entrants attack core business segments ¹. This practice surfaces assumptions, tests response capabilities, and builds organizational muscle memory for rapid adaptation.

Implementation Example: A search advertising platform conducts a two-day war game where one team role-plays an AI-native competitor launching a zero-click advertising model, while another team represents the incumbent response. The exercise reveals that current sales teams lack training on AI synthesis benefits and that product roadmaps don't include conversational interface development—critical gaps that weren't apparent in traditional planning processes. Following the war game, the organization accelerates conversational AI development by two quarters and implements sales enablement programs on AI positioning, directly informed by weaknesses exposed during simulation ¹.

Develop Hybrid Metrics Combining Traditional and AI-Era KPIs

Competitive intelligence frameworks should evolve beyond traditional metrics like market share and click-through rates to incorporate AI-era indicators including synthesis quality scores, conversational engagement depth, zero-click answer satisfaction, and cross-platform search diversification patterns ². This provides a more complete picture of competitive position as user behaviors shift.

Implementation Example: A market intelligence firm creates a "Search Disruption Index" combining five weighted metrics: traditional search market share (30%), AI platform query volume (25%), zero-click answer quality benchmarking (20%), voice/conversational search adoption (15%), and ecosystem partnership strength (10%). Monthly tracking shows their primary client's index score declining from 85 to 72 over six months despite stable traditional market share, revealing hidden erosion in AI-era competitiveness. This triggers strategic interventions including an OpenAI partnership and conversational interface development that wouldn't have been prioritized based on traditional metrics alone ¹².

Build Multidisciplinary Teams Combining AI Technical Expertise with Strategic Intelligence

Organizations should structure competitive intelligence teams to include data scientists with natural language processing expertise alongside traditional strategy analysts, enabling technical capability assessment rather than relying solely on market-facing indicators ³. This addresses the reality that AI disruption often manifests in technical capabilities before appearing in market metrics.

Implementation Example: A competitive intelligence unit adds two machine learning engineers who conduct technical teardowns of competitor AI search systems, analyzing model architectures, training approaches, and retrieval mechanisms through reverse engineering and academic publication analysis. This technical intelligence reveals that a competitor's recent accuracy improvements stem from a novel retrieval-augmented generation architecture that could be replicated, informing the organization's own technology roadmap. The multidisciplinary approach provides 6-9 month lead time on competitive capability shifts compared to waiting for market-visible impacts ³.

Implementation Considerations

Tool Selection and Technology Stack Integration

Implementing technological disruption risk monitoring requires careful selection of competitive intelligence tools that can handle both traditional web analytics and AI-specific capabilities including API-based testing, natural language quality assessment, and model performance benchmarking ¹³. Organizations must balance specialized AI monitoring tools with integration into existing CI platforms to avoid data silos.

For a mid-sized enterprise, this might involve combining Google Trends for traditional search pattern analysis, SimilarWeb for traffic benchmarking, custom Python scripts using OpenAI and Anthropic APIs for automated quality testing of competitor responses, and Tableau dashboards integrating all data sources. The key consideration is ensuring technical teams can maintain these systems as AI APIs evolve rapidly—one organization found their monitoring scripts broke monthly due to API changes, requiring dedicated engineering resources. Alternatively, emerging platforms like Reforge's AI Disruption Assessment tools provide integrated frameworks but may lack customization for industry-specific needs ¹.

Audience-Specific Customization of Intelligence Outputs

Competitive intelligence on technological disruption risks must be tailored differently for technical teams, strategic executives, and operational managers, as each audience requires different levels of technical depth and strategic framing ³. Technical teams need architectural details and capability comparisons, while executives require strategic implications and investment recommendations.

A practical implementation involves creating three intelligence product tiers: detailed technical teardowns for product and engineering teams (20+ pages with model architecture analysis), strategic briefings for executives (5 pages focusing on market implications and recommended actions), and operational playbooks for sales and marketing teams (tactical competitive positioning guidance). One organization found that providing executives with technical details led to decision paralysis, while giving sales teams only high-level strategy left them unable to address customer questions about AI capabilities. The tiered approach, updated monthly, ensures each audience receives actionable intelligence at appropriate granularity ³.

Organizational Maturity and Change Management

The effectiveness of technological disruption risk frameworks depends heavily on organizational readiness to act on intelligence, requiring assessment of AI literacy, risk tolerance, and decision-making speed before implementing sophisticated monitoring ¹. Organizations with low AI maturity may need foundational education before advanced competitive intelligence provides value.

A financial services firm discovered this when their comprehensive AI competitive intelligence reports sat unused because executives lacked context to interpret findings about retrieval-augmented generation or transformer architectures. They pivoted to a phased approach: first implementing AI literacy training for leadership, then introducing simplified competitive dashboards with clear strategic implications, and only later adding technical depth as organizational understanding matured. This change management approach, spanning 12 months, proved more effective than immediately deploying sophisticated intelligence to an unprepared audience. Success factors included executive sponsorship addressing the fact that 50% of leaders predict AI disruption yet many organizations rank it 18th among risks, revealing a dangerous awareness-action gap ⁶.

Resource Allocation and Build-vs-Buy Decisions

Organizations must determine whether to build custom technological disruption monitoring capabilities in-house or leverage external platforms and consultancies, considering factors including available technical talent, budget constraints, and required customization depth ¹³. This decision significantly impacts implementation timelines and ongoing maintenance requirements.

A retail company with strong data science capabilities chose to build custom monitoring using open-source tools and internal talent, achieving high customization for their specific competitive set (Amazon, Google Shopping, specialized AI shopping assistants) at lower ongoing cost but requiring 6 months development time. Conversely, a professional services firm lacking technical resources engaged a specialized CI consultancy providing AI disruption monitoring as a service, achieving faster deployment (6 weeks) but with less customization and higher ongoing costs. The key consideration is total cost of ownership over 3 years including maintenance, with build approaches favoring organizations with sustained technical resources and buy approaches better for those needing rapid deployment or lacking AI expertise ¹.

Common Challenges and Solutions

Challenge: Data Silos and Fragmented Intelligence Sources

Organizations struggle to integrate technological disruption signals from diverse sources including technical publications, user behavior analytics, competitor product releases, regulatory developments, and ecosystem partnership announcements ². This fragmentation creates incomplete pictures of competitive dynamics, with technical teams tracking model capabilities while strategy teams monitor market share, but neither connecting insights. The challenge intensifies as AI search competition spans traditional search engines, social platforms, retail sites, and specialized AI assistants, each requiring different monitoring approaches.

Solution:

Implement a centralized competitive intelligence platform with automated data aggregation from multiple sources and cross-functional access controls. Specifically, establish a cloud-based data warehouse (such as Snowflake or Google BigQuery) that ingests data from technical monitoring APIs, web analytics platforms, news aggregators, patent databases, and regulatory tracking services. Create unified dashboards using tools like Tableau or Power BI that present integrated views combining technical capability assessments, market metrics, and strategic implications. Assign a dedicated intelligence coordinator role responsible for synthesis across sources and facilitating weekly cross-functional reviews where technical, strategy, and operational teams share insights. One organization reduced intelligence fragmentation by 60% and improved decision speed by 40% through this centralized approach, discovering critical connections such as a competitor's AI model improvement correlating with a new university research partnership that neither technical nor strategy teams had connected independently ²³.

Challenge: Rapid Obsolescence of Competitive Assessments

Traditional competitive intelligence operates on quarterly or annual cycles, but AI capabilities can improve dramatically within weeks, rendering assessments obsolete before they inform decisions ¹³. For example, a detailed competitive analysis of ChatGPT's search capabilities completed in October 2023 became largely irrelevant when GPT-4 Turbo launched in November with significantly enhanced retrieval and synthesis. This creates a fundamental mismatch between intelligence production timelines and the pace of technological change.

Solution:

Transition from periodic comprehensive reports to continuous intelligence streams with automated monitoring and threshold-based alerting. Implement a "living document" approach where competitive assessments exist as continuously updated dashboards and wikis rather than static reports, with automated systems flagging significant changes. Establish clear thresholds for what constitutes material competitive shifts (e.g., >25% improvement in answer accuracy, >50% traffic growth, new capability launches) that trigger immediate alerts and rapid response protocols. Supplement automated monitoring with monthly "flash assessments" providing quick updates on key competitors rather than comprehensive annual reviews. One technology company implemented this approach using a combination of custom monitoring scripts, Slack alerts for threshold breaches, and bi-weekly 30-minute competitive intelligence standups, reducing average time from competitive shift to strategic response from 90 days to 12 days ¹³.

Challenge: Bias Amplification in AI Competitive Intelligence

AI-powered competitive intelligence tools themselves may introduce biases that distort strategic understanding, such as over-weighting easily quantifiable metrics while missing qualitative shifts in user preferences, or focusing on English-language developments while missing innovations in other markets ². Additionally, using AI tools like ChatGPT to analyze competitor AI capabilities creates circular dependencies where the analysis tool's limitations affect assessment quality.

Solution:

Implement multi-method triangulation combining AI-assisted analysis with human expert judgment, diverse data sources, and explicit bias checking protocols. Establish a practice of using multiple AI platforms (ChatGPT, Claude, Perplexity) for competitive analysis and comparing outputs to identify divergences that may indicate bias or limitations. Supplement automated analysis with quarterly expert panels including external advisors who provide alternative perspectives on competitive dynamics. Create explicit bias checklists reviewing whether intelligence disproportionately emphasizes certain geographies, user demographics, or capability dimensions. One competitive intelligence team discovered their AI-assisted monitoring was missing significant developments in Asian markets because their tools primarily indexed English-language sources; adding multilingual monitoring and regional expert consultants revealed competitive threats from Baidu and other platforms that weren't appearing in their assessments ².

Challenge: Executive Skepticism and Inaction Despite Clear Disruption Signals

Organizations often struggle to translate competitive intelligence on technological disruption into executive action, with leadership teams acknowledging AI disruption intellectually but failing to make corresponding strategic or investment decisions ⁶. Research shows 50% of leaders predict AI will change their business models, yet many organizations rank AI disruption 18th among risks and only 24% have secured their generative AI implementations, revealing a dangerous awareness-action gap.

Solution:

Reframe competitive intelligence presentations to emphasize financial impacts and strategic options rather than technical capabilities, and create forcing mechanisms that require explicit executive decisions on disruption responses. Translate technical competitive assessments into financial models showing revenue at risk under different disruption scenarios (e.g., "If zero-click searches reach 40%, our advertising revenue faces $X million annual impact"). Present intelligence alongside specific strategic options with clear resource requirements and expected outcomes, forcing choice rather than passive acknowledgment. Implement quarterly "disruption response reviews" where executives must explicitly decide to invest, monitor, or accept risks for each identified competitive threat. Use external validation through board presentations or third-party assessments to overcome internal skepticism. One organization broke through executive inaction by presenting competitive intelligence alongside a financial model showing $50 million revenue at risk over three years from AI search disruption, combined with three costed strategic options; this concrete framing led to approval of a $12 million AI integration initiative that had previously stalled despite clear competitive intelligence ⁶.

Challenge: Balancing Investment in Monitoring vs. Response Capabilities

Organizations face resource allocation dilemmas between investing in sophisticated competitive intelligence systems to detect disruption versus investing in AI capabilities to respond to disruption, with limited budgets forcing difficult tradeoffs ¹. Excessive investment in monitoring without corresponding response capabilities creates "analysis paralysis," while investing in AI development without competitive intelligence creates strategic blind spots.

Solution:

Adopt a portfolio approach that balances intelligence and response investments based on organizational maturity and competitive position, with explicit frameworks for resource allocation. For market leaders with significant resources at risk, allocate 20-30% of AI budgets to competitive intelligence and 70-80% to capability development, ensuring monitoring sophistication matches response capacity. For smaller players or new entrants, adopt a "fast follower" approach with lighter monitoring (10-15% of budget) focused on leading indicators and more investment in rapid capability deployment. Implement stage-gate processes where intelligence investments unlock corresponding response budgets—for example, detecting a competitive threat through basic monitoring triggers funding for deeper assessment and response planning. One mid-sized company optimized this balance by starting with lightweight monitoring using existing tools (Google Trends, free tiers of AI platforms, manual testing), which identified specific competitive threats; this intelligence then justified investment in both enhanced monitoring and AI capability development, creating a virtuous cycle where intelligence informed targeted responses rather than broad, unfocused AI investments ¹.

References

1. AlixPartners. (2024). The Future of Search: AI-Driven Disruption and Diversification. https://www.alixpartners.com/insights/102jze5/the-future-of-search-ai-driven-disruption-and-diversification/
2. CoinGeek. (2024). The Great Search Disruption: How AI, Gen Z Reshape Workplaces. https://coingeek.com/the-great-search-disruption-how-ai-gen-z-reshape-workplaces/
3. J.P. Morgan Asset Management. (2024). What Does AI Disruption Mean for Investors? https://am.jpmorgan.com/us/en/asset-management/adv/insights/market-insights/market-updates/on-the-minds-of-investors/what-does-ai-disruption-mean-for-investors/
4. IBM. (2024). 10 AI Dangers and Risks and How to Manage Them. https://www.ibm.com/think/insights/10-ai-dangers-and-risks-and-how-to-manage-them
5. Mercer. (2024). Managing People Risks in the Age of Technological Disruption. https://www.mercer.com/insights/total-rewards/employee-benefits-strategy/managing-people-risks-in-the-age-of-technological-disruption/
6. Reforge. (2024). AI Disruption Risk Assessment. https://www.reforge.com/blog/ai-disruption-risk-assessment