In the orchestra of modern commerce, the supply chain is the conductor’s baton. It determines rhythm, flow, and harmony. But in many firms, that baton is heavy, bloated with inefficiencies—warehouses stocked with slow-moving inventory, lead times stretching like rubber bands, and coordination that lags across partners.

Lean supply chain is the redesign: it is the commitment to eliminate waste, synchronize processes, and deliver value with precision. Like Toyota’s lean production, it anchors on doing more with less—less time, less inventory, less cost—while ensuring the customer always experiences more value.

A lean supply chain is not just about cutting fat; it is about improving muscular strength. Firms that adopt it optimize sourcing, automate logistics, and leverage data visibility to predict demand. They shift from reactive replenishment to proactive planning, ensuring that resources align with real consumption patterns.

In Africa, where logistics remains one of the biggest barriers to competitiveness, lean supply chain models can reduce inefficiencies that inflate prices and erode margins. Think of it as the engine that turns fragmented ecosystems into coherent pathways where goods move seamlessly and efficiently.

Ultimately, lean supply chain is both strategy and discipline. It demands that firms view partners not as adversaries to squeeze but as collaborators in a shared value chain. When businesses commit to lean systems, they free up capital, accelerate market responsiveness, and build resilience against shocks. The result is speed, adaptability, and sustained competitiveness.

From NATO today, we will be learning from a zen-master of lean at Tekedia Mini-MBA.

Thur, Sept 25 | 7pm-8pm WAT | Lean Supply Chain Applications in Business – Chibueze Noshiri, NATO Luxembourg | Zoom link

OpenAI, Oracle, and SoftBank have unveiled five new large-scale artificial intelligence data center sites across the United States under Stargate, OpenAI’s flagship infrastructure program.

The announcement, which cements the consortium’s role at the heart of America’s AI race, pushes the initiative to nearly 7 gigawatts of planned capacity and more than $400 billion in secured investments over the next three years.

With this milestone, Stargate is now on course to meet—and possibly surpass—its $500 billion, 10-gigawatt target by the end of 2025, ahead of the original schedule announced in January.

Expansion Across Texas, New Mexico, the Midwest, and Ohio

In July, OpenAI and Oracle deepened their partnership with a deal to deliver up to 4.5 gigawatts of new capacity. That agreement, valued at over $300 billion across five years, marked one of the largest corporate partnerships in AI infrastructure to date.

As part of this phase, three sites have now been selected: Shackelford County, Texas; Doña Ana County, New Mexico; and a soon-to-be-announced Midwest location. Alongside a potential 600-megawatt expansion at the flagship Abilene, Texas, Stargate campus, these facilities are projected to add more than 5.5 gigawatts of compute power. Collectively, they are expected to generate over 25,000 direct onsite jobs and tens of thousands of additional roles in supporting industries nationwide.

Meanwhile, SoftBank will spearhead two additional projects. In Lordstown, Ohio, the Japanese conglomerate has already broken ground on what it calls an advanced data center design, expected to go live in 2026. Another site in Milam County, Texas, will be delivered in partnership with SB Energy, SoftBank’s renewable energy subsidiary, which is building the powered infrastructure to support rapid deployment. Combined, the SoftBank sites will add up to 1.5 gigawatts within 18 months, with room for multi-gigawatt scaling.

The five sites were chosen from more than 300 proposals submitted by over 30 U.S. states following a nationwide competition launched in January. The rigorous selection process prioritized scale, energy infrastructure, and speed to deployment. Officials say further U.S. sites will be announced in the coming months as the partnership accelerates beyond its $500 billion goal.

Oracle, which is delivering cloud infrastructure for Stargate, confirmed that its Abilene campus—already operational on Oracle Cloud Infrastructure (OCI)—is seeing rapid progress. The company began delivering NVIDIA’s GB200 racks in June, marking a major leap in AI computing.

OpenAI has since started running early training and inference workloads on the system, signaling that the infrastructure is already feeding into next-generation model development.

Leaders Emphasize Scale, Speed, and Access

OpenAI CEO Sam Altman described compute infrastructure as the “key” to ensuring AI delivers benefits at scale.

“AI can only fulfill its promise if we build the compute to power it. That compute is the key to ensuring everyone can benefit from AI and to unlocking future breakthroughs. We’re already making historic progress toward that goal through Stargate,” Altman said.

Oracle’s Clay Magouyrk highlighted the speed of deployment, saying: “Oracle’s reliable, scalable, and secure AI infrastructure is helping OpenAI rapidly scale its business. To meet this enormous demand, we continue to expand OCI’s footprint at an unrivaled pace to deliver the most performant and cost-effective AI training and inferencing.”

Backdrop of U.S. AI Industrial Strategy

The Stargate initiative was first unveiled in January at the White House alongside President Donald Trump, framed as a cornerstone of the Administration’s drive to expand U.S. AI infrastructure and compete globally. Administration officials have credited Trump’s industrial policy framework for accelerating both domestic investment and foreign partnerships in the sector.

With Oracle, SoftBank, and other partners, OpenAI is translating its $500 billion promise into large-scale physical infrastructure that brings jobs, regional development, and computing capacity critical to powering the next wave of AI breakthroughs.

Executives say Stargate is not just about meeting current AI demand but building the foundation for a trillion-dollar future where compute availability advances artificial intelligence to advance humanity.

“Stargate is harnessing SoftBank’s innovative data center design and energy expertise to deliver the scalable compute that powers AI’s future. Together with OpenAI, Arm, and our Stargate partners, we are paving the way for a new era where AI advances humanity,” SoftBank Chairman, Masayoshi Son, said.

Uber CEO Dara Khosrowshahi has been blunt about what autonomous vehicles could mean for the millions of people who have built livelihoods on ride-hailing platforms.

Speaking at an All-In podcast summit, Khosrowshahi said human drivers will remain a major part of Uber’s network for the next five to seven years, but warned that “10 to 15 years from now, this is going to be a real issue,” adding that he has no “neat answer” for the social dislocation that could follow.

The company is already experimenting with autonomous options through partnerships such as its work with Waymo, but Khosrowshahi did not shy away from the human and political complexities ahead.

“For the next five to seven years, we’re going to have more human drivers and delivery people, just because we’re going so quickly,” he said. “But, I think, 10 to 15 years from now, this is going to be a real issue.”

He highlighted Uber’s efforts to create alternative on-demand work—labeling and other tasks to help train AI—as one way the company is trying to soften disruption, but acknowledged those roles require different skills than driving.

The Robotaxi Race: Who’s Launching, Who’s Partnering, and How They’re Trying to Scale

The industry is no longer hypothetical. Multiple firms are moving from labs and piloting to paying customers and app integration—often through deals with incumbent ride-hail apps—that aim to accelerate adoption while lowering unit economics.

Waymo, Alphabet’s autonomous-driving unit, is perhaps the farthest along in mainstream deployment. It now operates paid robotaxi services in several U.S. cities and has broadened coverage and product offerings aimed at business users, including a corporate bookings product that plugs Waymo into company travel programs. Waymo is expanding its footprint—now serving cities such as San Francisco, Phoenix, Los Angeles, Austin, and Atlanta—and completes more than a million autonomous rides per month, positioning itself as the most mature commercial robotaxi operator.

Rather than always building its own consumer brand, some AV developers are striking partnerships with large ride-hail platforms to gain immediate network scale. Waymo’s multi-city agreement to run its Waymo One fleet within the Uber app in selected markets is a clear example: Uber handles dispatching and payment while Waymo provides the fully driverless vehicles. That integration model lets robotaxi operators reach riders without having to build a consumer business from scratch.

Legacy automakers and start-ups have also formed alliances. Motional has run a long-standing commercial robotaxi service with Lyft, notably in Las Vegas, and maintains partnerships aimed at wider rollouts. The Motional–Lyft relationship exemplifies how AV companies are relying on incumbent ride-hail networks to reach real riders at scale.

New commercial tie-ups continue to surface. This month, a Lucid–Nuro–Uber arrangement took a visible step forward with the first delivery of a Lucid Gravity vehicle to be retrofitted for autonomous use, a milestone in plans to deploy thousands of robotaxis in the coming years under collaborations that involve vehicle makers, autonomy software firms, and ride-hail platforms. These industrial partnerships aim to solve manufacturing, retrofitting, and fleet economics in parallel.

Tesla remains a wild card. CEO Elon Musk has long promised a “robotaxi” future driven by Tesla’s Full Self-Driving (FSD) software. Tesla’s approach differs: instead of selling a standalone robotaxi fleet, Tesla plans a distributed model where consumer vehicles with FSD capabilities can be pooled as on-demand robotaxis. That model has sparked regulatory scrutiny and public confusion about what “robotaxi” means in Tesla’s terms—because many of Tesla’s public statements conflate supervised FSD features with fully driverless service.

Regulators have warned that supervised systems are not the same as true Level-4 robotaxis.

Risk, Regulation, and the Crash Test for Commercialization

Ambition runs headlong into regulatory reality. Cruise’s early charge into multiple cities and a rapid launch trajectory became a cautionary tale after safety incidents triggered permit suspensions and intense regulatory review. That episode showed how quickly public confidence and approvals can be jeopardized—forcing some operators to pause, re-test, and re-engage with regulators to restore trust and permissions.

Cities and state regulators are wrestling with how to certify and oversee robotaxis. Approvals vary widely across jurisdictions, and companies must navigate local requirements for testing, insurance, safety reporting, and human-supervision fallback strategies. That patchwork creates friction for scaling nationally and incentivizes partnership strategies—teaming with a local dispatch brand or proving reliability in a small set of cities first.

Even where regulators have green-lit services, city geography and user behavior shape the economics. Operators are focusing first on concentrated corridors—airport links, downtown commutes, and business travel—where predictable demand can be monetized while the systems accumulate safety data and learning. Waymo’s focus on airport connectivity and business accounts, for example, is a deliberate effort to capture high-value trips and predictable usage patterns.

Pilot data from commercial services suggest robotaxis can exceed many human drivers on efficiency metrics and utilization, but they still struggle with edge-case scenarios that human drivers manage easily—bad weather, unusual construction, and chaotic urban interactions. Some drivers told Business Insider they are skeptical that robotaxis can handle potholes and messy street conditions today; that skepticism helps explain why human drivers are not panicking yet. But efficiency gains from driverless fleets—if sustained—represent a structural pressure on labor economics for ride-hail platforms. (This dynamic was one of Khosrowshahi’s central points.)

What This Means for Uber and Drivers

For Uber, the short-term playbooks are straightforward: keep human drivers productive and pay them for demand that still requires human judgment; integrate robotaxi supply where regulators and partners allow; and expand alternative income streams on the platform (labeling, data work) to soften displacement.

“We’re expanding into other kinds of on-demand work as well to be able to adjust the kind of work available to people who want to earn on our own platform,” he said.

However, Khosrowshahi warned that there is no simple fix for the long run.

“This is a big, big societal question that we’re going to have to struggle with, and lots of others are going to struggle with too,” Khosrowshahi said.

In practice, that means Uber will run hybrid networks for years: human drivers handling most trips while robotaxis serve targeted corridors and scale gradually through partnerships (e.g., with Waymo) and vehicle supply deals. If robotaxis reach the scale their backers promise, the labor question moves from operational to societal—requiring public policy, reskilling programs, or new social supports to mitigate displacement.