How AI Will Kill CRM
Artificial intelligence is reshaping how enterprises operate and, in the not-too-distant future, may challenge long-held beliefs about corporations' customer relationships. Large organisations are pouring resources into AI-based solutions, reflecting the high-value executives place on data-driven insights, automation, and personalised experiences. From algorithms that optimise ad delivery to generative models that create deeply targeted content, AI-driven marketing technologies have become essential for businesses intent on growing their customer base and improving loyalty. Modern marketers deploy these tools at scale, weaving together chatbots, predictive analytics, and personalisation engines into a tightly integrated Martech stack that helps them fine-tune campaigns and boost conversion rates. Yet, a crucial question looms: are corporate leaders truly unlocking the disruptive power of AI, or merely grafting new systems onto old ways of doing business to chase short-term gains?
This question recalls a moment in history when the internal combustion engine was first invented. Some water-powered factories still relied on rivers that occasionally ran dry, so they used these new engines only to pump water back into reservoirs and keep water wheels turning. They never fully recognised that these engines could liberate their factories from depending on rivers altogether. This analogy highlights the danger of viewing AI as just another tool to preserve existing infrastructure. Many enterprises use machine learning to refine their marketing funnels or improve customer segmentation without questioning whether it simply prolongs outdated practices. Meanwhile, outside the walls of large corporations, entirely new AI paradigms are emerging that may ultimately eclipse today’s incremental strategies.
A powerful new wave of consumer-oriented AI now promises to upend traditional enterprise-customer relationships. Personal devices that people use every day are evolving into powerful localised engines of intelligence that empower individuals to take greater control over their data, interactions, and transactions. Apple Intelligence, for instance, delivers on-device AI capabilities that enable personal devices to make sophisticated predictions about user behaviour, automate daily tasks, and seamlessly integrate across multiple Apple offerings. Meanwhile, Google’s Project Astra is pathfinding a future where personal AI transcends the boundaries of any one device and becomes a service that follows users wherever they go, learning from their interactions across apps, sensors, and platforms. This personalised intelligence may soon become as ubiquitous as cloud-based infrastructure has become in enterprise computing. Meanwhile, in regions like Flanders in Belgium, innovative projects like Tim Berners-lee’s Inrupt personal data pods demonstrate how the web could evolve into a decentralised data-sharing ecosystem where people, rather than corporations, control access to their personal information. By rethinking who holds the keys to data, these pods create possibilities for decentralised privacy-preserving AI applications that perform sophisticated tasks on behalf of individuals, independent of or in coordination with the bigger corporate data pipelines.
This emerging wave of decentralised AI can be described as agentic AI, a concept that envisions a future in which personal AI agents act as trusted representatives for individuals, going beyond mere recommendation engines and actually engaging in actions traditionally performed by humans. Instead of just suggesting that users buy a particular product or subscribe to a specific service, agentic AI armed with digital IDs and verified credentials will be authorised to make these decisions on consumers' behalf. The shift from AI merely providing advice to AI executing tasks on behalf of its users is significant. It introduces a new model of interaction where humans transfer not just decision-making support to machines but also the authority to complete transactions, albeit often with the users “in-the-loop” providing oversite. This may herald a very different future – one in which agentic AI initiates negotiations and collaborates with other AIs to achieve goals set by its users. This level of delegation cuts to the core of what marketing, sales, and service design will look like in a future where businesses might not interface with humans at all.
How might ubiquitous agentic AI transform everyday life? Consider booking meetings, which has already started to become partially automated with tools like Calendly or AI-driven scheduling assistants. An agentic AI might not only schedule an appropriate slot on a calendar but also negotiate with other stakeholders’ AI assistants to find the ideal time and place, confirm the best available rates for travel, and coordinate last-minute changes. In the realm of utilities, an agentic AI could automatically monitor the market for better electricity or broadband deals, then seamlessly switch providers whenever it finds a plan that offers a balance of cost savings and service quality aligned with the consumer’s preferences. When planning a holiday, this personal AI might book flights, accommodations, and excursions, taking into account real-time data about weather patterns, crowd levels, and personal interests, and then finalise the transactions without requiring the user to fill out yet another form or call a travel agent. Even large logistical tasks like moving house could be simplified. Instead of an individual spending hours contacting moving companies, insurance providers, and local government offices, the agentic AI would gather quotes, compare rates, check regulatory requirements, and handle documentation, negotiating on the user’s behalf and finalising the best combination of services for the move. Each scenario reimagines the user not just as a customer, but as someone whose AI actively shapes the market by building new offerings from the fragmented services corporates offer. If corporates do not adapt, they might find themselves locked into outdated interactions with real people only at the final step—or cut out of that final step entirely if a competitor’s APIs align better with the personal agentic AI protocols.
For corporate marketing, these scenarios challenge every layer of the Martech stack. Today’s systems are designed to engage human consumers, even though much of the marketing process is automated on the enterprise side. Yet, corporate strategies must shift if personal AI agents become the gatekeepers of the individual’s data. Rather than crafting elaborate ads for human eyes, businesses will need to persuade AI agents whose primary goal is maximising the user’s defined preferences. Marketing teams accustomed to applying emotional appeals or brand narratives may find those tactics less effective when negotiating with algorithms that weigh quantifiable trade-offs. Further complicating things is the possibility that agentic AIs could outmanoeuvre even sophisticated corporate tech stacks. They might learn how to switch user behaviour quickly, exploit timing issues in dynamic pricing, or band together to negotiate bulk discounts. If millions of personal AIs join forces, they could dramatically flip market power away from incumbents.
Enterprises that assume humans will remain the main point of contact may discover their Martech funnels sidelined. Predictive analytics might still have value, but the data signals and feedback loops on which they rely will change fundamentally once AI agents control access. Without direct human touchpoints, brand loyalty could evaporate into purely algorithmic loyalty—if an AI perceives that rival services offer a better package, it can pivot instantly. Marketing materials might never reach human eyes and instead be filtered out by AIs that deem them irrelevant. On a larger scale, personal AIs might cooperate to exploit inefficiencies in corporate logistics, or en-masse time product purchases to off-peak demand or collaborate to gain crowd-based discounts. Such collective behaviour turns traditional digital marketing on its head and demands a rethinking of how companies architect their entire Martech stack.
This suggests that today’s corporate investments may mirror factories using combustion engines to pump water instead of fully leveraging a revolutionary invention. Although automating an existing marketing funnel can yield efficiency gains, it may miss the broader shift toward decentralised, agentic AI that grants individuals unprecedented leverage. Just as the internal combustion engine's real genius was its potential to unshackle factories from rivers, AI's true impact lies in its ability to empower individuals to manage their data, transactions, and relationships on their own terms.
The ramifications for Martech run deep. Much of marketing is currently geared toward capturing people’s attention, triggering emotional responses, or seeding brand recognition. In a future dominated by personal AIs, the emphasis shifts to forging AI-friendly interfaces, competitive pricing logic, and transparent value propositions that an algorithm can parse quickly. Gathering zero-party data becomes more complicated when individuals can easily shield it via decentralised data pods. Businesses that thrive in this environment will be those that rethink everything from product design to loyalty programs, ensuring their offerings meet the strict, data-driven criteria of agentic AI.
None of this is hypothetical; the pieces are already in play. Apple and Google are steadily pushing on-device AI, and decentralisation initiatives like Inrupt, Dataswyft, Dataspaien and Meeco are proving that individuals can hold and control their own data. Meanwhile, corporate AI strategies often centre on short-term wins—micro-targeting improvements, chatbots for customer support, automation or improved analytics pipelines. Though these initiatives can enhance efficiency, they fail to address how quickly consumer-facing AI could flip the script. If executives delay, they risk facing a radically altered marketplace where algorithmic intermediaries place or withhold orders, negotiate real-time pricing, and filter brand messages before humans ever see them. Strategies that cling to funnel-based marketing might eventually be as obsolete as water-powered factories in the age of abundant electricity.
The challenge, then, is to view AI not as a tool to make legacy models more efficient, but as a catalyst for ground-breaking change that will redefine how corporations and individuals interact. This moment calls for bold thinking about the role of data ownership, AI interoperability, and value exchange. Just as the internal combustion engine enabled entirely new industries—cars, airplanes, modern logistics—AI stands to transform business more deeply than most organisations are prepared for. Leaders who embrace this shift can create new forms of value by aligning with consumer-side agents, building trust at the algorithmic level, and finding ways to collaborate.
These transformative implications of the emergence of AI still lay largely untapped. Its onset invites a fresh wave of thinking: not how to embed AI into existing corporate infrastructures, but to understand how the nature of competition will change, how new corporate winners and losers will emerge and how new business models can sustainably create value. Make no mistake: the coming decade will see many established brands go to the wall while new market leaders emerge. Few enterprises have taken this notion seriously, partly because it challenges entrenched organisational practices and threatens near-term profits derived from an antiquated funnel-based approach. Yet the longer they wait, the more likely it becomes that their future markets will be defined by competitors who have embraced personal AI systems, leaving those who relied on incremental improvements locked out of a marketplace that has moved on. The challenge for corporate leaders is to start seeing personal agentic AI as a lever for radical transformation—one that will reshape not just the efficiency of operations but also the fundamental dynamics of how businesses engage with the people they aim to serve.