Editor’s note: This is AI Impact, Newsweek’s weekly newsletter where each week, we will explore how business leaders are unlocking real value through artificial intelligence.
Tap or click here to get this newsletter delivered to your inbox.
Signal Capture
Signals from the front lines of AI adoption
By Adam Mills
At a time when much of the AI conversation is centered on efficiency and cost-cutting, Calix CEO Michael Weening sees a gap between how AI is being talked about and how it’s actually being used inside companies.
“I call complete bullshit,” he told Newsweek, referring to claims that artificial intelligence is already driving large-scale workforce reductions. In his view, companies are using AI to justify cuts.
“There’s no way AI magically helped them eliminate 4,000 employees,” he said, pointing to recent examples of companies tying layoffs to AI adoption. “It sounds great. Investors get excited about it. But in the end, it’s just air cover.”
For Weening, the issue is not whether AI can improve efficiency (it can), but rather how companies are pursuing it. Framing AI primarily around cost reduction, he said, risks doing more harm than good and may lead companies to misread where value is created.
“It scares people,” he said. “And what it does is it seizes up the organization.”
At Calix, an AI and cloud platform company, that has meant taking a different approach—one that starts not with headcount reduction, but with how work actually gets done. Rather than layering AI tools on top of existing processes, the company has focused on rethinking workflows, asking employees to map out what they do today and identify where AI can augment or eliminate repetitive tasks.
Weening said employees across the company have built roughly 700 internal AI agents. Of those, about 40 have made it into broader production use.
“There’s a lot of AI stuff that’s going on which one would argue the value of,” he said.
Many of those agents, he said, are useful at an individual level—automating a specific task or improving a single workflow—but are not brought into broader enterprise use.
The examples that do move into production, he said, are those that have impact across teams or workflows.
Early deployments at Calix have focused on relatively contained use cases. In its IT organization, the company deployed an internal agent to handle common employee requests, such as password resets and ticketing. Within weeks, Weening said, the system delivered a 20 percent improvement in productivity.
In the company’s supply chain, AI agents have been used to analyze returns and error rates, surfacing patterns that would have previously required manual analysis. The goal, he said, is to give teams better insight into emerging issues so they can respond more quickly.
“What I would love to see is OpEx efficiencies,” he said. “Revenue goes up, but OpEx doesn’t go up at the same rate.”
That view comes as companies face increasing pressure from boards and investors to define their AI strategies. In many cases, Weening said, that pressure is warping expectations around what AI can deliver.
“Boards are watering at the mouth with regards to, ‘How do I go make all these cuts and be more profitable?’” he said.
At the same time, Weening said, leaders are being asked to deliver results on timelines that do not reflect how complex these kinds of transformations are.
“This is hard work,” he said. “It’s not like you sprinkle some fairy dust and great things happen.”
As companies move deeper into AI adoption, Weening said the risk is not just in how the technology is used, but how organizations respond to it. Focusing too narrowly on cost-cutting, he said, can undermine the broader effort to rethink how work is structured.
“I’ve never been able to cut my way to growth,” Weening said.
Core Intelligence
Tasks No Longer Define Work
The way companies define work is starting to change.
For years, work has been treated as a series of tasks—activities performed by people, supported by software and sold at scale. As AI systems begin to execute those tasks directly, that definition is breaking down.
“Work is a sequence of tasks,” said Tiger Tyagarajan, former CEO of Genpact and now a senior advisor and board member to a range of technology and services companies, as well as private equity and venture capital firms. “But the outcome of the work is what businesses care about—not the task.”
That distinction is becoming more important as the boundary between software and services begins to blur. Software has historically been designed to perform discrete functions, while services firms have sold human expertise to execute broader workflows. AI is collapsing that divide.
“AI tools today can perform tasks without human mediation, so the discussion shifts from tasks to work and outcomes,” said Dr. Ranjit Tinaikar, host of Newsweek’s “AI Impact Forum” series and the former CEO of Ness Digital Engineering.
As that shift takes hold, the unit of value begins to change. Companies are no longer optimizing for individual activities, but for the results those activities produce.
Delivering those results requires more than deploying a single tool. It requires coordinating systems.
“If you want to evolve to doing work, then you need orchestration,” Tyagarajan said.
That changes how work is organized. As execution becomes more automated, human roles move toward oversight—monitoring systems, intervening when needed and ensuring outputs meet expectations.
“It’s like driving a Tesla,” Tyagarajan said. “You’re not actually driving—but at various points, it might say, ‘Take control.’”
The implications extend to how services are priced. Models built on time and labor become harder to sustain when tasks can be executed faster and at lower cost.
“If you continue to charge on a time-and-materials basis, you likely will wind up on a path to zero,” he said.
That leads to a more fundamental question: What counts as work in the first place?
As AI takes on more of the execution layer, the line between tool and operator becomes less clear. What remains is not the task, but the outcome—and who is responsible for delivering it.
You can read the full article here: When AI Does the Work, What Changes?
Upcoming Webinars
How Is AI Reshaping Health Care Delivery and Strategy?
Health systems are entering a new phase of transformation as artificial intelligence moves from experimentation into everyday clinical and operational decision-making.
In this edition of Newsweek’s AI Agenda, “How Is AI Reshaping Health Care Delivery and Strategy?”, taking place Monday, April 27 at 3:30 p.m. Eastern, Suraj Srinivasan, Harvard Business School professor, sits down with Dr. Tom Mihaljevic, CEO and president of Cleveland Clinic, for a conversation on how one of the world’s leading health organizations is integrating AI into care delivery—and what that shift means for leadership priorities, operational strategy and the future of health care.
Register for free right now.
Will Today’s Software Companies Become Tomorrow’s Services Companies?
As artificial intelligence begins to take on more of the tasks traditionally performed by humans, a deeper question is coming into focus: what happens to “work” itself—and who, or what, delivers it?
In this session of Newsweek’s AI Impact Forum, “Will Today’s Software Companies Become Tomorrow’s Services Companies?”, taking place April 30 at 11 a.m. Eastern, Dr. Ranjit Tinaikar, host of the series, sits down with Tiger Tyagarajan, former CEO of Genpact and now a senior advisor and board member to technology and services companies, as well as private equity and venture capital firms, for a strategic conversation on how AI is blurring the line between software and services—and what that shift means for business models, competition and the future of work.
Register for free right now.
Upcoming AI Impact Forum Webinar
As AI accelerates across the enterprise, it is forcing leaders to rethink how technology spend is structured—and how much productivity it must deliver. In an upcoming “AI Impact Forum” session, Dr. Ranjit Tinaikar speaks with Noshir Kaka, senior partner at McKinsey & Company, about how this is reshaping enterprise spending—and what it means for organizations trying to keep up.
Register, for free, right now.
Prompt Injection
What’s one recent insight you’ve learned about AI?
“Trust is the final frontier for the merging of enterprise tech and AI. While we have seen generative and agentic AI solutions become significantly more refined since their inception, the risk of hallucinations still deters many leaders from fully embracing AI tools for decision-making. These leaders need guarantees, not guesses. And while AI alone cannot provide that, pairing it with mathematical optimization can.
By combining the predictive power of AI with the provable, mathematically based prescriptions of optimization, enterprises can deploy AI in mission-critical environments without sacrificing safety, quality or trust. This ultimately enables faster, better and more secure decision-making at every level, and enables us to move from ‘experimental’ to ‘mission-critical’ AI in 2026, deploying tools that enterprises can trust.”
Have your own lesson to share? Email us at: ai.newsletter@newsweek.com
Run Log
AI use case of the week
By Adam Mills
The workday doesn’t end when the last patient leaves. For many clinicians, documentation, prior authorizations and follow-up tasks often stretch into the evening—a block of after-hours work that many clinicians refer to as “pajama time.”
Fawad Butt, chief executive officer and co-founder of Penguin Ai, has seen how that burden builds across the system. As a former chief digital officer at Optum and Kaiser Permanente, he watched payer and provider organizations optimize their own processes, often adding friction that ultimately fell on physicians.
Penguin Ai was built to take on that administrative load. Its system handles execution tasks such as searching for clinical information, assembling patient data from multiple systems, interpreting payer requirements and managing authorization workflows. Each step is traceable, with clinicians staying in the loop rather than being removed from it.
What changes is how time is used. Work that once spilled into evenings can be absorbed into the day, while decisions are supported by more consistent and complete information. The effect is less about automation and more about removing the layers of process that separate clinicians from patient care.
Butt said the difference is often only clear in hindsight. “Physicians don’t realize how worn down they are until something changes,” he said. “When the repetitive work is gone, you stop filling out checklists and start thinking again.”
His takeaway is straightforward. AI doesn’t replace the clinician’s role; it clears the administrative barriers that have come to define it, allowing physicians to focus on the parts of care that require judgment and experience.
Have an AI use case to share with us? Email us at: ai.newsletter@newsweek.com
Context Window
■ Sullivan & Cromwell apologized to a federal judge after submitting a court filing with AI-generated errors, including fabricated legal citations, in a bankruptcy case, and later filed a corrected motion after the issue was flagged by a rival law firm. [Reuters]
■ Apple’s incoming CEO John Ternus faces mounting pressure to define the company’s artificial intelligence strategy, as the tech giant lags rivals in generative AI and relies on partners like Google and OpenAI to power key features. [CNBC]
■ An AI agent has been deployed to run a retail store in San Francisco, handling tasks like hiring and inventory, but early results have been uneven, with poor product selection and scheduling issues highlighting the limits of autonomous systems. [The New York Times]
■ Sen. Jon Ossoff has launched an inquiry into whether the rapid expansion of AI data centers is contributing to rising electricity costs, asking federal regulators how they will ensure technology companies do not pass those costs on to consumers. [CBS News]
■ Jeff Bezos–backed startup Project Prometheus is raising about $10 billion in new funding at a roughly $38 billion valuation, as the company focuses on “physical AI” systems designed for real-world industrial applications like manufacturing and aerospace. [Business Insider]
Transfer Protocol
Tracking executive moves across the AI landscape
Sameer Gupta, previously managing director and chief analytics officer at DBS Bank, has been appointed chief data and AI officer at Lloyds Banking Group, where he will lead the bank’s artificial intelligence strategy and oversee its adoption across the business while establishing governance and responsible use frameworks.
Jason Luo, previously chief revenue officer at Actabl and a former chief executive officer at ALICE, has been appointed chief executive officer at Newo, where he will lead the company’s growth and expansion of its AI voice platform through a partner-led go-to-market strategy.
Elizabeth Smalley, previously chief product officer at Marigold, has been appointed chief AI officer at Marigold, where she will lead global AI and data strategy and integration of intelligent capabilities across its product suite and operations.
Axelle de Faÿ, previously a customer success and transformation leader at Salesforce, has been appointed chief customer experience officer at Sidetrade, where she will oversee the integration of AI across customer operations and delivery.
Andrew Shaw, previously director of product at OLX and a former senior product leader at adidas, has been appointed chief product and technology officer at Candid, where he will lead product strategy, technology infrastructure and the scaling of its AI-powered platform.
Know someone on the move in AI? Send job change info to ai.newsletter@newsweek.com
Magic Moment
What’s the most fun or unexpected way you’ve used AI lately?
“I’m a venture capitalist by day, but I also produce electronic music under the name VTLX. I’m currently making an album with an audio engineer in Australia, singers scattered across three continents and session musicians I’ve never met in person. The hardest part isn’t the music—it’s making sure everyone is building toward the same vision when no one is in the same room.
I used AI to turn my rough ideas—reference tracks, tempo notes, mood descriptions, even voice memos—into structured creative briefs for every collaborator. What would have taken me a full weekend of writing took an afternoon. Each brief spoke that person’s language: technical specs for my engineer, emotional direction for vocalists, groove references for musicians.
What stood out is that AI didn’t make the music—it made the collaboration possible at a level of clarity I couldn’t have pulled off alone. As someone who evaluates AI startups for a living, I talk about the technology’s potential every day. But using it to conduct a global creative project from my home studio made the impact personal. AI’s quiet superpower isn’t replacing people—it’s connecting them.”
Experience some AI magic? Tell us about it at ai.newsletter@newsweek.com
Tap or click here to get this newsletter delivered to your inbox.
Read the full article here
