StarZero All shows Try the podcast clipper →

Jul 9, 2026 · 22m · big-technology

The Real ROI of AI Tokens — With Dallas Dolen

Dallas Dolen · 15m spoken Alex Kantrowitz · 5m spoken
0:00 / 0:00
▶ Watch on YouTube →

gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions

Alex Kantrowitz sits down with PwC's Dallas Dolen to explore the financial and operational realities of enterprise AI adoption, focusing on token budgeting, governance control planes, and shifting from vanity metrics to outcome-driven ROI.

How this conversation actually went

Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Alex holds 25.8% of the talking time here. How this is scored →

Alex as informed peer 4.8 Guest teaching 4.0 Guest disagreement 1.3 Alex pushing back 2.5
05100:0010:0020:000:54–4:31 · Alex as informed peer 6/10 Debating the Reality of Enterprise Token Maxing Alex frames the discussion with real enterprise cases like Uber's rapid budget depletion and Meta's cost-cutting before asking whether token minimizers or maxers will win. Dallas reframes the dichotomy, arguing the real metric of enterprise success is outcome maximization rather than token spend.4:31–8:52 · Alex as informed peer 5/10 Measuring Enterprise AI ROI and Work Product Quality Alex presses on whether cloud foundation labs intentionally obscure token tracking to maximize enterprise spend. Dallas explains how enterprise control planes enforce governance and prevent wasteful use cases, comparing overpowered models on routine tasks to driving a Lamborghini to buy milk.8:53–11:07 · Alex as informed peer 5/10 Audience Polling on AI Pricing and Model Price Wars Alex conducts a live show-of-hands poll on willingness to pay double for AI capabilities and brings up potential model price cuts by OpenAI. Dallas analyzes the audience reaction, noting an observable drop in enterprise willingness to absorb price hikes compared to months earlier.11:07–16:41 · Alex as informed peer 6/10 Model Price Sensitivity in Large-Scale Organizations Alex asks how major price cuts would alter deployment and presses on the operational limits of autonomous agents. Dallas details how PwC manages high price sensitivity across 350,000 employees and categorizes agent limitations into risk, cost, and organizational decision-making tolerances.16:46–19:00 · Alex as informed peer 2/10 Generational Heritage and the Human Purpose of Technology Alex respectfully offers condolences on the passing of Dallas's grandmother and invites him to share her story. Dallas reflects on his family's generational history in San Francisco and ties it to building technology that genuinely improves human lives.19:01–22:40 · Alex as informed peer 5/10 Human Augmentation vs. Relinquishing Control to Agents Alex asks how enterprise leaders become comfortable losing control when deploying autonomous agents in high-stakes environments. Dallas challenges the loss-of-control premise, arguing that agents act as human augmentation rather than an unmonitored transfer of authority.0:54–4:31 · Guest teaching 4/10 Debating the Reality of Enterprise Token Maxing Alex frames the discussion with real enterprise cases like Uber's rapid budget depletion and Meta's cost-cutting before asking whether token minimizers or maxers will win. Dallas reframes the dichotomy, arguing the real metric of enterprise success is outcome maximization rather than token spend.4:31–8:52 · Guest teaching 5/10 Measuring Enterprise AI ROI and Work Product Quality Alex presses on whether cloud foundation labs intentionally obscure token tracking to maximize enterprise spend. Dallas explains how enterprise control planes enforce governance and prevent wasteful use cases, comparing overpowered models on routine tasks to driving a Lamborghini to buy milk.8:53–11:07 · Guest teaching 4/10 Audience Polling on AI Pricing and Model Price Wars Alex conducts a live show-of-hands poll on willingness to pay double for AI capabilities and brings up potential model price cuts by OpenAI. Dallas analyzes the audience reaction, noting an observable drop in enterprise willingness to absorb price hikes compared to months earlier.11:07–16:41 · Guest teaching 6/10 Model Price Sensitivity in Large-Scale Organizations Alex asks how major price cuts would alter deployment and presses on the operational limits of autonomous agents. Dallas details how PwC manages high price sensitivity across 350,000 employees and categorizes agent limitations into risk, cost, and organizational decision-making tolerances.16:46–19:00 · Guest teaching 1/10 Generational Heritage and the Human Purpose of Technology Alex respectfully offers condolences on the passing of Dallas's grandmother and invites him to share her story. Dallas reflects on his family's generational history in San Francisco and ties it to building technology that genuinely improves human lives.19:01–22:40 · Guest teaching 4/10 Human Augmentation vs. Relinquishing Control to Agents Alex asks how enterprise leaders become comfortable losing control when deploying autonomous agents in high-stakes environments. Dallas challenges the loss-of-control premise, arguing that agents act as human augmentation rather than an unmonitored transfer of authority.0:54–4:31 · Guest disagreement 2/10 Debating the Reality of Enterprise Token Maxing Alex frames the discussion with real enterprise cases like Uber's rapid budget depletion and Meta's cost-cutting before asking whether token minimizers or maxers will win. Dallas reframes the dichotomy, arguing the real metric of enterprise success is outcome maximization rather than token spend.4:31–8:52 · Guest disagreement 1/10 Measuring Enterprise AI ROI and Work Product Quality Alex presses on whether cloud foundation labs intentionally obscure token tracking to maximize enterprise spend. Dallas explains how enterprise control planes enforce governance and prevent wasteful use cases, comparing overpowered models on routine tasks to driving a Lamborghini to buy milk.8:53–11:07 · Guest disagreement 1/10 Audience Polling on AI Pricing and Model Price Wars Alex conducts a live show-of-hands poll on willingness to pay double for AI capabilities and brings up potential model price cuts by OpenAI. Dallas analyzes the audience reaction, noting an observable drop in enterprise willingness to absorb price hikes compared to months earlier.11:07–16:41 · Guest disagreement 2/10 Model Price Sensitivity in Large-Scale Organizations Alex asks how major price cuts would alter deployment and presses on the operational limits of autonomous agents. Dallas details how PwC manages high price sensitivity across 350,000 employees and categorizes agent limitations into risk, cost, and organizational decision-making tolerances.16:46–19:00 · Guest disagreement 0/10 Generational Heritage and the Human Purpose of Technology Alex respectfully offers condolences on the passing of Dallas's grandmother and invites him to share her story. Dallas reflects on his family's generational history in San Francisco and ties it to building technology that genuinely improves human lives.19:01–22:40 · Guest disagreement 2/10 Human Augmentation vs. Relinquishing Control to Agents Alex asks how enterprise leaders become comfortable losing control when deploying autonomous agents in high-stakes environments. Dallas challenges the loss-of-control premise, arguing that agents act as human augmentation rather than an unmonitored transfer of authority.0:54–4:31 · Alex pushing back 3/10 Debating the Reality of Enterprise Token Maxing Alex frames the discussion with real enterprise cases like Uber's rapid budget depletion and Meta's cost-cutting before asking whether token minimizers or maxers will win. Dallas reframes the dichotomy, arguing the real metric of enterprise success is outcome maximization rather than token spend.4:31–8:52 · Alex pushing back 3/10 Measuring Enterprise AI ROI and Work Product Quality Alex presses on whether cloud foundation labs intentionally obscure token tracking to maximize enterprise spend. Dallas explains how enterprise control planes enforce governance and prevent wasteful use cases, comparing overpowered models on routine tasks to driving a Lamborghini to buy milk.8:53–11:07 · Alex pushing back 2/10 Audience Polling on AI Pricing and Model Price Wars Alex conducts a live show-of-hands poll on willingness to pay double for AI capabilities and brings up potential model price cuts by OpenAI. Dallas analyzes the audience reaction, noting an observable drop in enterprise willingness to absorb price hikes compared to months earlier.11:07–16:41 · Alex pushing back 4/10 Model Price Sensitivity in Large-Scale Organizations Alex asks how major price cuts would alter deployment and presses on the operational limits of autonomous agents. Dallas details how PwC manages high price sensitivity across 350,000 employees and categorizes agent limitations into risk, cost, and organizational decision-making tolerances.16:46–19:00 · Alex pushing back 0/10 Generational Heritage and the Human Purpose of Technology Alex respectfully offers condolences on the passing of Dallas's grandmother and invites him to share her story. Dallas reflects on his family's generational history in San Francisco and ties it to building technology that genuinely improves human lives.19:01–22:40 · Alex pushing back 3/10 Human Augmentation vs. Relinquishing Control to Agents Alex asks how enterprise leaders become comfortable losing control when deploying autonomous agents in high-stakes environments. Dallas challenges the loss-of-control premise, arguing that agents act as human augmentation rather than an unmonitored transfer of authority.

speaking balance: gold is Alex, purple is the guest (3 minute bins)

0:00 · Alex 51.6% · guest 48.4%0:00 · Alex 51.6% · guest 48.4%3:00 · Alex 6.1% · guest 93.9%3:00 · Alex 6.1% · guest 93.9%6:00 · Alex 14.3% · guest 85.7%6:00 · Alex 14.3% · guest 85.7%9:00 · Alex 53.8% · guest 46.2%9:00 · Alex 53.8% · guest 46.2%12:00 · Alex 16.9% · guest 83.1%12:00 · Alex 16.9% · guest 83.1%15:00 · Alex 7.1% · guest 92.9%15:00 · Alex 7.1% · guest 92.9%18:00 · Alex 29.9% · guest 70.1%18:00 · Alex 29.9% · guest 70.1%21:00 · Alex 28% · guest 72%21:00 · Alex 28% · guest 72%
Sharpest disagreement ▶ 1:09 Disagreeing on token maxing being pure media hype

Dallas directly dissents from Aaron Levie's claim that token maxing is merely a media invention, stating that problematic spending behaviors in enterprise environments are very real.

 Hardest push from Alex ▶ 6:27 Challenging cloud labs on opaque token billing

Alex challenges the guest by raising reports that major foundation model providers deliberately structure integrations to maximize token burn while obstructing granular ROI measurement.

 Biggest teaching moment ▶ 11:30 Explaining scale elasticity across 350,000 workers

Dallas educates the host on how enterprise price sensitivity functions at scale across 350,000 employees, detailing automated control-plane routing that restricts premium model access by default.

 Alex holds their own ▶ 1:54 Citing enterprise budget burnouts at Meta, AT&T, and Uber

Alex displays sharp industry knowledge by citing specific cost containment strategies at AT&T and Meta alongside Uber burning through its full-year AI budget in months.
the scores for every segment, with the reasoning behind each
ChapterTopicAlex as informed peerGuest teachingGuest disagreementAlex pushing backWhy
Debating the Reality of Enterprise Token Maxing 6423 Alex frames the discussion with real enterprise cases like Uber's rapid budget depletion and Meta's cost-cutting before asking whether token minimizers or maxers will win. Dallas reframes the dichotomy, arguing the real metric of enterprise success is outcome maximization rather than token spend.
Measuring Enterprise AI ROI and Work Product Quality 5513 Alex presses on whether cloud foundation labs intentionally obscure token tracking to maximize enterprise spend. Dallas explains how enterprise control planes enforce governance and prevent wasteful use cases, comparing overpowered models on routine tasks to driving a Lamborghini to buy milk.
Audience Polling on AI Pricing and Model Price Wars 5412 Alex conducts a live show-of-hands poll on willingness to pay double for AI capabilities and brings up potential model price cuts by OpenAI. Dallas analyzes the audience reaction, noting an observable drop in enterprise willingness to absorb price hikes compared to months earlier.
Model Price Sensitivity in Large-Scale Organizations 6624 Alex asks how major price cuts would alter deployment and presses on the operational limits of autonomous agents. Dallas details how PwC manages high price sensitivity across 350,000 employees and categorizes agent limitations into risk, cost, and organizational decision-making tolerances.
Generational Heritage and the Human Purpose of Technology 2100 Alex respectfully offers condolences on the passing of Dallas's grandmother and invites him to share her story. Dallas reflects on his family's generational history in San Francisco and ties it to building technology that genuinely improves human lives.
Human Augmentation vs. Relinquishing Control to Agents 5423 Alex asks how enterprise leaders become comfortable losing control when deploying autonomous agents in high-stakes environments. Dallas challenges the loss-of-control premise, arguing that agents act as human augmentation rather than an unmonitored transfer of authority.

Statements from this episode (5)

Assertion Supported
Uber exhausted its entire annual AI budget in under six months
“And we also know that Uber, of course, spent their entire budget in less than half a year.”
Alex Kantrowitz Jul 9, 2026 ▶ 2:07
Assertion Not checkable as stated
PwC automatically routes employees to specific AI models based on use case
“And we've even built that into some of our control plane technology. So there's selectivity of model and there's also recommendations within it as well. So depending on who you are and what it is that we think you're going to do with it automatically configure…”
Dallas Dolen Jul 9, 2026 ▶ 12:20
Disclosure
PwC is shifting its hiring focus toward science and non-engineering graduates
“We're hiring as many people as we did last year, and we're changing the who we're hiring. We're actually hiring a lot more kids who are pursuing sciences and, you know, even in areas that they go beyond, you know, engineering.”
Dallas Dolen Jul 9, 2026 ▶ 15:41
Disclosure
PwC is hiring fewer people for pure accounting roles
“I think we're hiring a little bit in the less of just like, you know, pure accounting. Also a function of where we're doing it. It's not that we're hiring less people overall. We're hiring less people in certain areas.”
Dallas Dolen Jul 9, 2026 ▶ 15:54
Insight
Enterprise AI agent adoption limits depend on error, cost, and governance tolerance
“The limits are going to be a tolerance for error, a tolerance for costs and a tolerance Again, for like what will be acceptable within the organization of having that thing do versus having, you know, a human do it.”
Dallas Dolen Jul 9, 2026 ▶ 16:19
Made with StarZero

Turn any episode into a week of clips.

This entire site, over 300 episodes transcribed, diarized, checked and made playable, runs on the StarZero media pipeline. Drop in your own episode and the podcast clipper finds the moments worth sharing, cuts them, captions them, and reframes them for every feed.

Try the podcast clipper Learn more