Decoder with Nilay Patel: "Why IBM CEO Arvind Krishna is still hiring humans in the AI era"

Episode Date: December 1, 2025
Guest: Arvind Krishna, CEO of IBM
Host: Nilay Patel, The Verge

Episode Overview

In this engaging episode, Nilay Patel sits down with Arvind Krishna, CEO of IBM, for a candid discussion about IBM’s evolving role in the tech industry, its big bets on artificial intelligence (AI) and quantum computing, and how Krishna views the tumultuous, hype-driven moment technology is experiencing. The conversation dives deep into why IBM is still hiring humans when competitors are making layoffs, the company’s lessons from Watson, the reality behind AI’s economics, the future of quantum, and how IBM is navigating waves of technological change.

Key Discussion Points and Insights

1. IBM’s Evolution and Focus (06:03–09:00)

• B2B Transformation: Arvind describes IBM’s shift over decades from consumer-oriented products (typewriters, PCs) to being fully an enterprise, B2B technology provider focused on deploying tech that improves clients’ businesses.
• Watson’s Legacy: The Watson AI computer was pivotal in putting AI on the public map, but its move into healthcare was “inappropriate” and too monolithic for market needs.
  • “We were trying to be too monolithic and we picked maybe one of the toughest areas, healthcare... that was inappropriate.” – Arvind Krishna (07:29)

2. The AI Technology Bets: Then & Now (09:00–14:03)

• Watson vs. LLMs: IBM’s early forays with Watson used cutting-edge machine learning and early deep learning models but suffered from being bespoke and inflexible, compared to today’s modular, scalable LLMs.
  • “Right technology, wrong go-to-market approach.” – Arvind Krishna (09:31)
• LLMs as the Great Unlock: LLMs changed the economic equation, reducing the labor and time needed for AI deployment by 100x, making it “industrial scale.”

3. Cost of AI and the Role of GPUs (14:03–18:33)

• Skyrocketing Costs: Despite LLMs dropping some costs, current AI scale requires massive hardware investments (mainly in Nvidia GPUs), leading to high CapEx.
• Future Efficiencies: Arvind is optimistic about a coming “10x” reduction in costs from improvements in semiconductors, AI chip designs, and software—translating to up to 1,000x cost savings over five years.
  • “If you put those three tens together, that’s 1,000 times cheaper... even if you get the square root of that, that’s 30 times cheaper for the same dollar to be spent.” (16:46)

4. Bubbles, Booms, and Tech Cycles (18:33–29:14)

• Is AI a Bubble? Krishna doesn’t believe it is, but predicts not all capital invested will see ROI; analogizes to the fiber boom—initial overinvestment led to lasting infrastructure, though not all investors won.
  • “We spend the money, it gets corrected back to 30 cents on the dollar. At that point, it makes an incredible amount of sense for somebody else to get that asset and turn it into a profit stream.” (19:35)
• Hardware vs. Infrastructure: Unlike “eternal” assets like fiber, GPUs and hardware investments will depreciate quickly, but Krishna points out that much of the data center infrastructure will have enduring value even as chips change.

5. Centralization, App Economies, and Platform Dynamics (29:14–35:36)

• AI is echoing smartphone-era dynamics where ecosystems (like App Stores) became choke points that collected “rent” on economic activity.
  • “On the back of that transition [to apps], Apple and Google... collected a huge amount of fees. They are some of the richest companies in the world.” – Nilay (32:19)
• Krishna sees room for disruption yet, noting that innovation can shift and that competitive platforms could arise, especially in enterprise AI. Not all control points are locked down.

6. IBM’s Own Restructuring and Bet on Quantum (35:36–47:41)

• Hybrid Cloud & Red Hat: Krishna’s pivotal bet, exemplified by the Red Hat acquisition (30% of IBM’s market cap), paid off and reoriented IBM as a hybrid cloud leader.
  • “If that conviction turned out to be wrong, I should be fired. That is why I keep the red hat as a reminder...” (40:29)
• Three Big Bets: Focus on hybrid cloud, enterprise AI, and quantum computing. IBM exited commodity businesses to concentrate on high-innovation areas.
  • “Our sweet spot is helping our B2B clients succeed... what are we really good at? We're really good at building systems. And so the third bet that I decided early on is we're going to make a bet on Quantum.” (35:58)

7. Quantum Computing’s Promise (47:41–54:47)

• Early Stage, Lasting Investment: IBM’s quantum strategy is to engage developers and enterprises now (via open-source software, research partnerships) to be ready when utility-scale quantum emerges.
  • “If 650,000 [developers] had been 1,000, I would have told my people, guys, this is your physics friends. This is not a market.” (51:33)
• Quantum’s Potential: Quantum will not replace, but augment CPUs/GPUs (like GPUs did for CPUs)—Krishna expects significant, defensible value to accrue for early movers.
  • “QPU’s are going to have an incredible value because they can solve problems you actually cannot solve on GPUs and CPUs...” (52:47)

8. AI, AGI, and the Limits of Current Technology (54:47–66:25)

• Sober Approach vs. Hype: Krishna sharply contrasts IBM’s numerically driven, measured investments in quantum and enterprise AI against the speculative consumer AGI playbook (e.g., OpenAI).
  • “There’s no way you’re going to get a return on [current AI infra spend] is my view... It’s a belief that one company is going to be the only company that gets the entire market.” (59:25)
• LLMs’ Limitations: LLMs are not the endpoint; the future of AI will require new architectures, including integration of hard knowledge and possibly new technologies wholly distinct from LLMs.
  • “I give it really low odds... that the current set of known technologies gets us to AGI.” (60:28)

9. AI and the Workforce: Why IBM is Still Hiring (66:25–71:50)

• AI as Force Multiplier: IBM’s internal experience shows AI (specifically, AI-coding tools) increases productivity (by 45%) for programmers, leading Krishna to double down on hiring technical talent rather than cutting jobs.
  • “Wouldn’t I rather take an entry-level person and the AI makes them more like a 10-year expert? Isn’t that more useful for me than the other way around?” (67:35)
• Job Market Outlook: Krishna expects up to 10% displacement in certain job categories, but sees this as part of natural business cycles and believes that increased productivity will spur growth elsewhere.

10. Looking Ahead for IBM

• Quantum as Near-Term Focus: Big breakthroughs in quantum computing expected within 2–3 years.
  • “Watch what we're going to do on Quantum. I think that in about two to three years you'll see some surprising results.” (71:54)
• Continued Growth Orientation: IBM will keep pushing innovation, hiring technical talent, and staking its own path instead of chasing hype.

Notable Quotes & Memorable Moments

“We were trying to be too monolithic and we picked maybe one of the toughest areas, healthcare, to go into, which I think was inappropriate.”
— Arvind Krishna (07:29)

“Right technology, wrong go-to-market approach.”
— Arvind Krishna (09:31)

“If you put those three tens together, that’s 1,000 times cheaper...even if you get the square root of that, that’s 30 times cheaper for the same dollar to be spent.”
— Arvind Krishna (16:46)

“We spend the money, it gets corrected back to 30 cents on the dollar. At that point, it makes an incredible amount of sense for somebody else to get that asset and turn it into a profit stream.”
— Arvind Krishna (19:35)

"If that conviction turned out to be wrong, I should be fired... That is why I keep the red hat as a reminder to myself."
— Arvind Krishna (40:29)

“QPU’s are going to have an incredible value because they can solve problems you actually cannot solve on GPUs and CPUs in any economic terms in the near term.”
— Arvind Krishna (52:47)

“I give it really low odds... that the current set of known technologies gets us to AGI.”
— Arvind Krishna (60:28)

“Wouldn’t I rather take an entry level person and the AI makes them more like a 10 year expert? Isn’t that more useful for me than the other way around?”
— Arvind Krishna (67:35)

“We are hiring more because...if I can be that much more productive at software development, that means that we can build a lot more products, which means we can go get more market share. It doesn’t mean that it’s a fixed amount of work. I think the amount of work is infinite so we can be more productive.”
— Arvind Krishna (70:46)

Important Timestamps

• [06:19] – IBM’s B2B identity, shift from consumer to enterprise
• [07:29] – Watson’s impact and missteps
• [09:31] – Watson’s initial tech, LLMs, modularity
• [11:20] – Differentiating LLMs versus earlier deep learning
• [16:42] – Cost reduction potential in AI hardware/software
• [19:35] – AI bubble analogy, betting on infrastructure
• [32:19] – App store centralization, AI’s parallel risks
• [35:58] – IBM’s three big bets: hybrid, enterprise AI, quantum
• [40:29] – Red Hat as symbol of conviction and change
• [51:33] – 650,000 quantum software developers: proof of traction
• [54:47] – Quantum as “and” (not replacement) to chips
• [60:28] – AI spend vs. ROI; AGI skepticism
• [67:35] – AI and job displacement; IBM’s hiring philosophy
• [71:54] – What’s next: quantum in 2-3 years

Tone & Approach

The interview exudes a blend of measured optimism and technical rigor, with Krishna offering candid reflections on IBM’s missteps and successes. His answers are grounded in historical awareness and a strategic, “slow and steady” worldview, consistently advocating for sustainability, ongoing innovation, and prudent risk-taking. The tone is thoughtful, considered, and often skeptical of hype, with both host and guest reacting to current tech enthusiasm with sober, data-driven analysis.

Summary Takeaway

IBM’s CEO Arvind Krishna sees a future where AI and quantum live side by side, not in a world of mass layoffs but a transformation of work and technology. With pragmatic lessons from Watson, deep bets on quantum, and a focus on making enterprise clients successful, IBM is carving its own path—even as the larger tech world chases moonshots and hype cycles. Krishna’s healthy skepticism, bet on hiring over layoffs, and patient investment in new foundational technologies offer a sobering—yet hopeful—template for tech leadership in the AI era.