In the news. As AI adoption accelerates, electricity—not models or chips—is emerging as the next strategic constraint, according to Harvard Business Review. Data centers are expected to consume more than twice as much electricity by 2030 as they do today, making access to reliable power a competitive differentiator. With companies investing heavily in AI, the article’s authors argue that leaders should make energy central to their AI strategies. This includes securing long-term energy options, optimizing where computing runs, and carefully measuring and managing power needs. [HBR]

As global economic pressures intensify, productive investment offers a practical way through, serving as a proxy for competitiveness and a gauge of productivity growth.

On McKinsey.com. Energy is just one of the changing dynamics reshaping global competitiveness, making investment decisions more consequential and complex than ever. New research from the McKinsey Global Institute argues that where and how organizations invest is defining future competitiveness. McKinsey’s Anna Kortis, Jan Mischke, Chris Bradley, Shubham Singhal, Sylvain Johansson, and their colleagues share how organizations can strengthen their positions by evaluating investment decisions at the project level, weighing the factors that matter most for each investment, securing energy access, and prioritizing locations where innovation ecosystems flourish.

Catalyze competitive advantage

In the news. Many AI initiatives succeed in controlled pilots but falter when companies attempt enterprise-wide deployment, Fortune reports. Industry leaders say the challenge is less about getting the technology right than about governance, process design, and organizational alignment. Another frequent obstacle is fragmented data and unclear workflows, which can delay deployment. Generating value from AI requires strong direction from leaders, as well as cross-team coordination, defined business goals, and a focus on metrics—not just successful proofs of concept. [Fortune]

On McKinsey.com. Successful AI rollouts are only as strong as their data foundations, according to McKinsey experts Asin Tavakoli, Brian Goodman, Kayvaun Rowshankish, Stephen Reddin, and coauthors. The authors find that only 7% of companies have fully scaled AI. A big barrier? Data that’s not ready. Two-thirds of high-performing organizations identify data as a primary roadblock to scaling AI. But it’s not just about needing better quality data. To scale AI with trust and consistency, organizations need to treat data as a shared enterprise asset, building reusable data foundations that support AI applications across the business.

Build strong data to scale AI

Rewiring drug development

In the news. Pharmaceutical companies are boosting AI investments as they seek to improve the economics of drug development, The Wall Street Journal reports. Roughly 90% of drug candidates fail, so companies are using powerful computing and gen AI to identify promising therapies faster. Early evidence suggests AI can shorten research timelines and improve manufacturing efficiency, but industry experts caution that evidence of improved clinical success rates is still emerging. Nonetheless, AI is becoming central to the search for new medicines. [WSJ]

On McKinsey.com. AI’s greatest impact on biopharma R&D will come from redesigning how decisions are made, according to McKinsey experts Alex Devereson, David Champagne, Erika Stanzl, Lieven Van der Veken, and their coauthors. AI-powered “learning loops” could transform traditional drug development by connecting discovery, molecule design, clinical trials, patient outcomes, and commercialization pathways. Used this way, AI can continually improve research decisions, helping companies reduce uncertainty, compress development timelines, and bring medicines to market faster.

Transform R&D with AI