The day in one read

The Revaluation of Design in the Age of Agentic AI

The narrative surrounding artificial intelligence’s impact on the creative and technical workforce has shifted from existential dread to a more nuanced debate over agency, cost, and operational reality. While public sentiment suggests that designers are the most unhappy demographic in technology, industry leaders argue that the current period of uncertainty is actually a unique opportunity for the profession. This tension is not merely cultural; it is structural. As AI tools move from experimental novelties to core infrastructure, companies are grappling with the economic realities of token-based pricing, the need for rigorous governance, and the subtle erosion of human skills. The day’s dominant story is not that AI is replacing humans, but that it is fundamentally altering the economics of labor, forcing a re-evaluation of how value is created, measured, and distributed across the enterprise.

The Economics of Intelligence and Token Costs

A significant divergence exists between the theoretical promise of AI productivity and the financial reality of deploying it at scale. Traditional software operates on a one-time investment model, but AI introduces a recurring operating expense driven by marginal costs per use. This shift has caught many organizations off guard, with firms reporting that they have exhausted annual AI budgets within months. The result is a new discipline of financial management centered on "token budgets" and usage caps. Sarah Fryer, CFO of OpenAI, highlighted this transition in the context of building an AI-native finance function. She noted that 40% of finance professionals’ specialized AI use involves non-finance work, while 22% involves engineering tasks, suggesting that AI is blurring functional boundaries. Fryer recommends that organizations measure "value per unit of intelligence" rather than simply tracking token consumption, a metric that aligns cost with actual business output.

This economic pressure is reshaping executive strategy. Greg Shove, CEO of Section, advises CEOs to avoid the binary trap of "token maxing" or "token minimizing." Instead, he suggests identifying a "lighthouse team" where investment can be increased tenfold to drive disproportionate value. This approach contrasts with the broader trend of cost-cutting, suggesting that strategic concentration is more effective than blanket reduction. The implication is that AI is not a uniform utility to be applied everywhere, but a variable resource that requires careful allocation. Companies that treat AI as a line item to be minimized may miss the opportunities for high-leverage applications, while those that spend without governance risk financial instability. The emerging consensus is that AI costs must be managed with the same rigor as any other major operational expense, requiring new financial frameworks and KPIs.

The Myth of Immediate Productivity and Job Loss

Despite the hype surrounding AI’s transformative potential, evidence suggests that the productivity boom is neither immediate nor uniform. EY has identified three major misconceptions regarding enterprise AI adoption, the first being the belief that AI generates an instant productivity surge. Historical parallels indicate that infrastructure build-out and talent development typically precede macroeconomic gains. Organizations are currently experiencing "jagged" productivity gains as they integrate agentic workflows with human oversight, a process that is often slower and more complex than anticipated. This reality check is crucial for investors and executives who may be modeling overly optimistic growth scenarios based on early adopter successes.

The second misconception is that AI will render labor redundant. Claims that 40% or more of recent layoffs are attributable to AI are dismissed by EY as convenient excuses, with evidence of rehiring undermining this narrative. The data suggests that while AI is changing the nature of work, it is not yet eliminating the need for human labor at a scale that would drive mass unemployment. Instead, the risk lies in "distributed de-skilling," a concept highlighted by BCG Global Chair Rich Lesser. BCG research indicates that half of surveyed leaders see this phenomenon, with over 60% expecting it to be a threat within three to five years. Lesser argues that the risk most leaders are not tracking is not AI hallucinations or job loss, but the gradual erosion of human expertise as workers rely increasingly on AI for routine tasks. This de-skilling could undermine the very human oversight that is necessary to manage AI systems effectively, creating a feedback loop of declining competence.

The Future of Design and Human Agency

In the face of these operational challenges, the role of design is undergoing a profound revaluation. Ian Silber, Head of Product Design at OpenAI, argues that the current uncertainty regarding design roles stems from a lack of clarity on expectations, not a decline in value. While sentiment surveys indicate that designers are unhappy, Silber contends that this is the best time in history to be a designer due to unprecedented agency. He notes that while engineers have seen 10x to 100x productivity gains from AI, the design process remains fluid and iterative, requiring human feedback loops that AI cannot yet fully replace. At OpenAI, the design approach varies by feature: some ChatGPT components undergo rigorous testing where 99 out of 100 ideas are discarded, while others, like Codex, embrace rapid, public iteration. This duality reflects the broader tension between precision and speed in AI development.

Silber emphasizes that AI is already an incredible product designer, accessible to all, but humans remain essential for understanding unmet needs, inventing new interaction paradigms, and providing a distinct point of view. He predicts a shift in startup ratios, potentially moving from one designer to 15 engineers to two designers to one engineer, as design becomes a primary differentiator. This inversion of the traditional engineering-heavy model suggests that as AI handles more of the technical implementation, the ability to define the problem and shape the user experience becomes the critical bottleneck. The future of ChatGPT, according to Silber, involves a "super app" with proactive capabilities, richer voice inputs, and adaptive interfaces that simplify the user experience. This vision requires designers who can balance the needs of billions of casual users with the complex requirements of power users, aiming for a universal input that seamlessly handles diverse tasks.

Corporate Culture and the Fight Against "AI Slop"

The integration of AI into daily workflows is also prompting a cultural reckoning within companies. Varun Anand, co-founder of Clay, instituted a company-wide AI writing policy to combat what he terms "AI slop." The policy mandates that employees stand behind every sentence, treats writing as thinking, requires more time spent writing than consuming, and rejects length as a proxy for quality. This approach reflects a broader concern that the ease of AI-generated content is leading to a dilution of intellectual rigor and accountability. By requiring employees to take ownership of their output, Clay is attempting to preserve the human element of communication and decision-making.

This cultural shift is part of a larger trend toward "AI-native" operations, where the goal is not just to use AI tools, but to restructure workflows around them. OpenAI’s Sarah Fryer outlined ambitions for a "zero-day close" and automated forecasting in finance, illustrating how AI can enable new operational models. However, these models require a workforce that is not only technically proficient but also culturally aligned with the principles of accountability and critical thinking. The challenge for leaders is to balance the efficiency gains of AI with the need for human judgment and creativity. This requires a deliberate effort to define what it means to be an "AI-native" organization, including clear policies on content creation, decision-making, and skill development.

The Strategic Shift to Ecosystem Commitment

As enterprises move past the hype cycle, the focus of executive inquiry is shifting from model selection to ecosystem commitment. BCG Global Chair Rich Lesser notes that CEOs are increasingly asking how to build an "enterprise cortex" to preserve proprietary knowledge. This concept suggests that the value of AI lies not just in the models themselves, but in the integration of those models with a company’s unique data, processes, and culture. The risk of "distributed de-skilling" is particularly acute in this context, as companies that outsource too much of their intellectual work to AI may lose the internal expertise needed to maintain and improve their systems.

KPMG’s Adaptability Report adds another dimension to this strategic shift, stating that executives are twice as likely to invest in technology as in employee training. However, the data suggests that this imbalance may be a mistake. 37% of leaders who increased workforce investment reported revenue rises of 20% or more, compared to 25% overall. This correlation implies that human capital development is a key driver of AI-driven growth, not a secondary concern. The implication is that companies that invest in both technology and talent are better positioned to capture the value of AI than those that focus solely on the former. This finding challenges the prevailing narrative that AI is a pure technology play, suggesting instead that it is a holistic transformation that requires a balanced approach to investment.

Also Notable

The broader landscape of AI adoption is also being shaped by the emergence of new tools and platforms that are redefining the boundaries of software development. Lovable, Notion, and other platforms are enabling non-technical users to build complex applications, further blurring the line between design and engineering. This trend supports Ian Silber’s prediction of a shift in startup ratios, as the barrier to entry for creating software products continues to lower. Meanwhile, the automotive and mobility sectors are also feeling the impact of AI, with companies like Rivian and Waymo leveraging advanced algorithms to improve safety and efficiency. These developments suggest that the influence of AI is extending far beyond the tech sector, touching every industry that relies on data-driven decision-making.

In the realm of corporate governance, the need for robust AI oversight is becoming a board-level priority. Companies are establishing new roles and committees to manage AI risk, ensuring that their use of these tools aligns with ethical and legal standards. This is particularly important as AI systems become more autonomous, with the potential to make decisions that have significant real-world consequences. The challenge for leaders is to strike a balance between innovation and responsibility, ensuring that AI is used in ways that benefit society as a whole. As the technology continues to evolve, the need for clear guidelines and best practices will only grow, making governance a critical component of any successful AI strategy.