https://www.facebook.com/167107520008024/

The Baruch Group, San Marcos, CA (2026)

06/23/2026

Seeing real value from AI depends on being able to verify its outputs
bySeb Murray
Jun 8, 2026
What you’ll learn:

AI can now produce work faster than humans can verify it.
Closing that gap is becoming the central economic challenge of the transition to artificial general intelligence.
Firms that can go beyond AI deployment to underwrite the risks of AI outputs will have a competitive advantage.
Artificial intelligence can now generate thousands of lines of code in minutes, and many companies are taking advantage of that capability. However, what AI cannot yet do reliably at scale is ensure that its code is safe, correct, or complete.

Humans try to fill that gap, reviewing outputs line by line, though that is becoming less workable as systems produce more code than any individual can realistically audit. Often, the code ships anyway.

The gap between fast AI output and slower human verification is at the heart of a new paper by MIT Sloan School of Management research scientist Christian Catalini and his co-authors, Xiang Hui of Washington University and Jane Wu, SM ’18, PhD ’22, of UCLA.

In “Some Simple Economics of AGI,” the researchers lay out an economic model of the transition toward artificial general intelligence — AI systems that can operate with broad autonomy across many tasks. The researchers focus on the problem of measurability: whether the outputs of those systems can be reliably checked.

As AI systems are becoming more capable, it’s getting harder to verify everything they produce. This will put a cap on how fully the benefits of AGI can be realized in the economy: AI makes it cheap to produce work, but not to judge whether that work is any good.

Verification will become a core part of seeing value from AI
AI has largely been considered as a substitute for labor, with the assumption that cheaper output can translate directly into value. But what will distinguish firms is not so much their ability to deploy AI as their ability to stand behind what it produces, the researchers write.

The gap between what AI systems can produce and what can be properly checked is widening. On the SWE-bench AI performance benchmark, the accuracy of AI coding tools rose from 4.4% to 71.7% in a year, and the length of tasks that systems can complete is doubling over short periods, according to the researchers. But there is “scarce capacity” for human verification, as bandwidth continues to be constrained by time and experience, the researchers write.

Verifying AI outputs is no longer just a compliance function but a core part of how value is created from AI, the researchers write. That puts a premium on records of how systems behave — especially where they fail — and on taking responsibility when things go wrong.

Catalini describes this as a shift from software as a service to what he calls “liability as a service.” “The companies that understand the risks and can underwrite them will be the ones that profit,” he said.

The risks of using AI to verify AI and skipping verification altogether
So far, AI adoption has clustered in areas where outputs can be checked quickly: summarizing text, generating images, writing code. But as systems take on longer, higher-risk tasks — and as AI agents act autonomously — checking whether they were done correctly will become more difficult and often take more time, increasing the risk of misplaced trust.

One response from companies is to use AI to check AI, which the researchers call a “tempting shortcut.” But where both systems share the same assumptions, they can reinforce the same errors, creating what Catalini described as a false sense of confidence and not a real solution.

For some companies, competitive pressure and the gap between fast, cheap production and slower verification creates the motivation to deploy systems before they are fully checked by humans, according to the researchers, allowing risks to build unnoticed until they are harder to contain.

But when systems are pushed into use before they are fully verified, the consequences can be severe. Catalini pointed to episodes such as the 2010 flash crash in financial markets, where complex automated systems failed in ways that were not fully understood at the time. “If we do not invest in verification, we’re accumulating hidden risk,” Catalini said. “It is technical debt accumulating behind the scenes, and, at some point, it’ll come due.”

In effect, the economy becomes “hollow,” the researchers write: Output surges, but the quality and utility of the output don’t keep pace. The researchers describe this as a “Trojan horse” problem, where unverified output leaks into the economy and is treated as if it were reliable.

A new paper explores how seeing economic value from artificial intelligence hinges on closing the gap between what AI can do and how humans can verify its outputs.

06/23/2026

Global Economics Intelligence executive summary, April 2026
May 29, 2026 | Article

Summarize
Oil price increases impact production costs and household budgets, while growth loses momentum; nevertheless, the broader economy continues to expand, leaving interest rates in a holding pattern.

The Middle East conflict continues to dominate global energy markets. The closure of the Strait of Hormuz has disrupted a route that normally carries around 20 million barrels of oil per day and more than one-quarter of global seaborne oil trade. The UAE’s May 1 decision to leave OPEC adds another layer of uncertainty: While the immediate market impact may be constrained by the Hormuz disruption, the move could weaken OPEC’s longer-term market position.

The energy strain is currently feeding through to the broader commodity complex. Oil and gas prices remain elevated and volatile, with fertilizer and metals prices also coming under pressure. This matters because the shock is no longer only about energy supply; it is becoming a cost shock for producers and consumers. Manufacturing firms are facing higher fuel, transport, and input costs. Emerging economies are particularly exposed because higher energy prices often pass through into food prices, import costs, and household purchasing power more quickly. S&P Global noted that emerging-market manufacturing costs rose sharply in March as the Middle East conflict lifted fuel, transport, commodity, and dollar-denominated import prices.

Real GDP in the US increased at an annual rate of 2.0% in the first quarter of 2026, up from the 0.5% posted in the fourth quarter of 2025. This real GDP increase reflects upturns in government spending and exports—and an acceleration in investment, partly offset by a deceleration in consumer spending. China was even more buoyant, reporting a resilient growth rate of 5.0% year on year for the first quarter of 2026, outpacing the fourth quarter of 2025 (4.5%). In contrast, global uncertainty has resulted in lowered GDP growth forecasts for the eurozone: Oxford Economics now expects just 0.8% this year (down by 0.2 percentage points compared with March), while the International Monetary Fund anticipates 1.1%.

Consumers may feel under siege: Higher prices are eroding real incomes while uncertainty is weighing on confidence. Retail sales and consumer spending may still look resilient in nominal terms—especially where energy prices lift total spending values—but the actual picture is weaker. In real terms, higher fuel and food prices are likely crowding out discretionary consumption, especially in energy-importing and lower-income economies. Nevertheless, in the US, the Consumer Confidence Index edged up to 92.8 in April from an upwardly revised 92.2 in March.

Nominal retail sales in March increased due to higher oil prices but, in real terms, overall consumer spending slowed significantly. In the US, March retail and food services sales in the US (adjusted for seasonal variation and holiday and trading-day differences) were $752.1 billion, down 1.7% from February’s revised $739.8 billion.

Inflation expectations are rising, especially at the five-year outlook, which reached almost 3%—levels last seen in 2022–23. March saw US median inflation expectations at the one-year-ahead horizon rise to 3.4% (from 3.0%), while expectations at the three-year-ahead horizon increased to 3.1%.

Consumer inflation has accelerated across the board, driven primarily by higher energy prices (Exhibit 1). Notably, consumers in emerging economies faced higher prices, driven not only by higher energy costs but also by rising food prices. Among developed economies, the US Consumer Price Index (CPI) was up 3.3% year over year in March, after rising 2.4% in February; core inflation rose 2.6% (annualized). Similarly, consumer price inflation in the eurozone climbed sharply to 2.6% annually in March, the highest rate since July 2024. The rise was almost exclusively driven by higher fuel prices, even after some countries implemented tax cuts to cushion consumers from the impact. Among emerging economies, retail inflation in India rose to 3.40% in March, a ten-month high under the new CPI series (base year 2024), and up from 3.21% in February. The rise reflected higher fuel costs and the West Asia conflict premium, especially in energy and transport, with food inflation firming to 3.87% after prior months of contraction.

Updated monthly in 2026, the McKinsey Global Economics Intelligence executive summary offers detailed insights, trends, and analysis on global trade.

06/22/2026

From Reactive Recovery To Predictive Resilience: The AI Shift In Business Continuity
ByGouri Sankar Dash,Forbes Councils Member.
for Forbes Technology CouncilCOUNCIL
Jun 10, 2026,

Gouri Sankar Dash, Engagement Director at TCS with 20+ years driving enterprise data, AI platforms and multimillion-dollar transformations.

The Limits Of Traditional BC/DR Models
Traditional business continuity and disaster recovery (BC/DR) models struggle to keep pace with modern enterprise demands. Designed for a slower, less complex world, these frameworks often rely on manual intervention and assume that recovery windows of several minutes are acceptable. In today’s real-time economy, however, even brief downtime can result in significant financial loss, regulatory exposure and reputational damage.

What is emerging is a fundamental shift: from reactive recovery to predictive resilience.

AI As The Engine Of Transformation
Artificial intelligence is central to enabling this transformation. By continuously analyzing patterns across infrastructure, applications and transaction flows, AI allows organizations to detect anomalies early and act before disruptions escalate. Instead of responding after failures occur, businesses can proactively mitigate risks.

Equally important is AI’s ability to drive automated response. Traditional systems often introduce delays due to human decision making. AI-enabled environments, by contrast, can instantly isolate affected systems, trigger failover mechanisms and maintain operational continuity with minimal interruption. This capability is particularly critical in sectors such as finance, where milliseconds can have a material impact.

At a broader level, organizations are moving toward adaptive resilience frameworks—systems that learn, evolve and improve over time. Unlike static playbooks, AI-driven models continuously refine their response strategies based on historical incidents and emerging threat patterns. Resilience, therefore, becomes embedded within the organization’s operational DNA.

Research That Backs The Shift
This direction is increasingly reinforced by both academic research and industry practice. Governance, risk and compliance expert Ramachander Rao Thallada has examined how AI-driven business continuity can reduce downtime through predictive intelligence, automated response and faster recovery in financial services. His work highlights a key point: Traditional continuity models are becoming insufficient in dynamic, high-risk environments.

The same theme appears in broader cyber-resilience thinking. Strategic cybersecurity leaders such as Vikram Das have emphasized that cyber resilience is now foundational to business continuity, especially as organizations face growing exposure across identity, cloud, endpoints and backup environments. This practitioner view is important because resilience is no longer limited to disaster recovery teams; it now requires executive visibility, risk ownership and business-aligned decision making.
Together, these perspectives point to the same conclusion: The future of BC/DR will not be defined only by how quickly organizations recover, but by how early they can detect risk, how intelligently they can coordinate response and how continuously they can learn from disruption.

Building A Foundation For AI-Enabled Resilience
However, implementing AI-enabled resilience goes beyond deploying advanced algorithms. Organizations must invest in integrated data ecosystems, robust technology architectures and strong governance frameworks. Transparency and explainability are especially important in regulated industries, where AI systems must justify their decisions and actions.

Resilience As Competitive Advantage
Ultimately, the future of business continuity will not be defined by how quickly organizations recover, but by how effectively they predict and prevent disruption. This shift from reactive to proactive resilience represents one of the most significant transformations in modern enterprise strategy.

Organizations that embrace this evolution can not only reduce risk but also gain a sustainable competitive advantage. In an unpredictable digital landscape, resilience is no longer a safeguard; it is a differentiator.

The transition from traditional recovery models to intelligent, AI-driven resilience marks a defining moment in business strategy. Enterprises that harness data, automation and adaptive systems will be better positioned to navigate uncertainty while maintaining operational continuity and customer trust. Early adopters will not only minimize disruptio

This shift from reactive to proactive resilience represents one of the most significant transformations in modern enterprise strategy.

06/22/2026

How Elite Sports Coaches Make High-Pressure Decisions
by Alan McCall, Adrian Wolfberg, Johann Bilsborough and Ricard Pruna
From the Magazine (forthcoming July–August 2026)

Kyle Ellingson
Summary. Drawing on interviews with 11 elite coaches across major U.S., European, Australian, and New Zealand professional sports leagues, the authors show that high-pressure decisions aren’t a single “formula” but a set of practices that unfold before, during, and after pivotal moments—and translate directly to business leadership. Top coaches combine disciplined preparation with emotional control and social awareness during crunch time and accountability and continual system improvement after the fact.close

Business leaders routinely make important decisions under pressure, often with incomplete or conflicting information, in ways that significantly impact team and organizational performance—as well as their own careers. Elite sports coaches do the same, with two critical factors increasing the stakes: a compressed time frame (they often must call plays in just seconds) and constant public exposure (including live TV coverage and 24/7 criticism from fans and the media).

Over the past several years we have studied 11 successful coaches working in the National Football League (NFL), National Basketball Association (NBA), and Major League Baseball (MLB) in the United States; Premier League, LaLiga, and UEFA Champions League (football) and Rugby Union in Europe; and the National Rugby League in Australia and New Zealand to better understand how they handle everything from in-game play calls and substitutions to recruitment and return-from-injury choices. We wanted to understand what they did cognitively, emotionally, and socially as they made decisions.

What emerged was not a checklist of traits or a new decision formula. Instead, we were able to carefully document how high-stakes decisions take shape before, during, and after moments of consequence. In this article we focus on specific practices the coaches employ during each of those three phases—and explain how business leaders can use them to increase the quality of their own decision-making.

Drawing on interviews with 11 elite coaches across major U.S., European, Australian, and New Zealand professional sports leagues, the authors show that high-pressure decisions aren’t a single “formula” but a set of practices that unfold before, during, and after pivotal moments—and translate...

06/22/2026

Risk and resilience in the AI era: More data doesn't mean better decisions
By Sanchita Chakraborti
At 2:13 AM, a vulnerability scanner flags critical exposure in a customer-facing application. Within minutes, other signals begin surfacing across the environment. A cloud security platform detects configuration drift in a Kubernetes cluster.

An observability tool shows latency spikes in a dependent API service. Infrastructure monitoring flags resource saturation in a production environment. At the same time, network operations teams begin investigating intermittent packet loss affecting users in a specific location.

Every team has data. Every dashboard is functioning correctly. Every alert is technically accurate. And yet, no one can answer the most important operational question:

What matters right now?

This is the reality of modern enterprise operations. Organizations are not struggling because they lack visibility. Over the past decade, they have invested heavily in observability, monitoring and security tools. Most can detect nearly everything happening across their environments. But visibility alone does not create resilience.

Beyond detection: Cutting through the noise
The real challenge is turning those fragmented signals into a shared understanding—and then into coordinated action fast enough to reduce operational risk before it disrupts the business. This is where resilience is now won or lost.

For years, enterprises have focused on improving detection capabilities. As environments became more distributed and cloud-native architectures introduced new layers of complexity, organizations responded by deploying more tools, including observability platforms, vulnerability scanners, SIEM systems and real-time telemetry pipelines.

Those investments worked. Applications generate telemetry continuously. Infrastructure emits signals across every layer. Security platforms identify vulnerabilities at scale. Cloud environments surface exposure paths instantly.

Organizations have more visibility than ever before. The problem is that detection has now outpaced coordination, and AI is accelerating this imbalance. Modern AI-driven systems can identify vulnerabilities, insecure dependencies, configuration drift and exposure paths at a scale far beyond what human teams can manually triage or remediate. This method creates a new kind of operational pressure.

The challenge is no longer finding vulnerabilities. It is determining which exposures matter, which systems are at risk, what will impact the business and what should be addressed first.

Without clear answers, teams become overwhelmed by noise. And noise is the enemy of resilience.

Because resilience is not about collecting more signals. It is about understanding what those signals mean together, prioritizing risk based on real business impact and remediating vulnerabilities at scale before disruption spreads.

Organizations don’t just need better prioritization. They need the ability to automate and orchestrate remediation across fragmented environments, security tools, IT operations workflows and infrastructure domains. Without automation, even the right insights become bottlenecks.

True resilience comes from connecting signals across domains, turning insight into coordinated action and reducing the time between detection, decision and remediation.

Lack of coordination: Where resilience breaks down
Operational failures rarely happen because a single alert is missed. They happen because organizations cannot connect information across teams fast enough to act effectively.

Consider a retail enterprise preparing for a major holiday sales event.

Security teams identify a high-severity middleware vulnerability affecting part of the commerce stack. At the same time, infrastructure teams are scaling systems for traffic surges. Application teams are deploying final promotional updates. Network teams are troubleshooting latency issues affecting checkout performance.

Individually, every team is operating correctly. But without shared context, decision-making becomes fragmented and risky.

The vulnerability flagged as critical might affect a low-priority service, while a lower-severity issue in a payment dependency could pose far greater business risk. A rushed fix could destabilize already stressed systems, but a delayed response could expose customers during peak revenue hours.

This is where resilience breaks down. Not because of missing tools or lack of expertise, but because operations remain fragmented across disconnected systems and workflows. Each team sees a different version of reality.

Resilience depends on bringing those perspectives together into one coordinated model.
The goal is not simply to detect issues faster. It is to create shared operational context across teams, prioritize actions based on business impact, and enable coordinated remediation before risks become disruptions.

Why modern resilience depends on turning AI-driven discovery into coordinated action

06/22/2026

Your budget is killing your strategy—here are four ways to fix it
May 29, 2026 | Article
By Matthew Maloney
with Deidre Harrison and Michele Tam

Summarize
Traditional budgeting can undermine performance. Today, leading CFOs use budgets to turn strategy into action.

Every year, companies invest thousands of hours in building their budgets, only to find within months that the plan no longer matches reality. Demand shifts, input costs change, competitors reposition, and new technologies reset productivity baselines. Yet the budget locks capital into allocations based largely on last year’s choices.

Summing up the problem, a former e-commerce CFO said, “If you stick to a rigid budget, you often find that by the time it’s approved, the world has already moved on, and you’re left trying to catch up.”

The uncomfortable truth is that the traditional annual budget process is not just inefficient; in today’s environment, it is strategically corrosive. In addition to being backward-looking, it reinforces incrementalism and can make dynamic reallocation politically and operationally difficult.

However, a handful of companies are outperforming their peers on growth, resilience, and total shareholder returns, and they’re contributing to their success not by budgeting better, but by budgeting differently. They treat the budget not just as a way to control spending, but as a road map for executing strategy. They have evolved the budgeting process in four ways: they base budgets on strategic choices and value-creation potential, they shift 10 to 20 percent of capital year over year toward higher-return opportunities, they replace single-scenario budgeting with scenario-based planning, and they use AI and machine learning to turn operational data into forward-looking insights.

Providing further evidence of how fundamentally companies are rethinking budgeting, a small number have gone even further by eliminating traditional budgets altogether in favor of approaches built around targets, real-time reporting, automated forecasting, and more dynamic performance management. While still the exception, these approaches underscore the extent to which the role of budgeting is being reconsidered.

This article explores traditional budget building and why a new approach is needed. It then explores the four primary ways top companies are reinventing the process and suggests action steps for CFOs with the courage to redesign the system.

The four imperatives of the new budgeting process

Historically, the annual budget served a critical purpose. In high-growth environments, it created structure, alignment, and basic financial discipline. As the e-commerce CFO recalled about working with earlier-stage companies, the “extremely important” yearly process “was long and cumbersome, like two to three months.”

For scaling organizations, this kind of budgeting provides necessary discipline. The former CFO of a technology services firm recalled that during a period of rapid growth, the company needed to build more structure into its operations and financial management.

“When I first started, we were very immature from an infrastructure perspective. We were growing at 50 or 60 percent a year. As the business got larger and more complicated, we needed to put more structure in place,” she said. The annual budgeting process was part of that structure, helping impose financial discipline as the company scaled.

However, structure is not strategy.

Leading CFOs are rethinking annual budgets, using budget planning to connect strategy, rolling forecasts, and dynamic capital allocation.

06/21/2026

IDEAS MADE TO MATTER ARTIFICIAL INTELLIGENCE
Seeing real value from AI depends on being able to verify its outputs
bySeb Murray
Jun 8, 2026 5 minute read
What you’ll learn:

AI can now produce work faster than humans can verify it.
Closing that gap is becoming the central economic challenge of the transition to artificial general intelligence.
Firms that can go beyond AI deployment to underwrite the risks of AI outputs will have a competitive advantage.
Artificial intelligence can now generate thousands of lines of code in minutes, and many companies are taking advantage of that capability. However, what AI cannot yet do reliably at scale is ensure that its code is safe, correct, or complete.

Humans try to fill that gap, reviewing outputs line by line, though that is becoming less workable as systems produce more code than any individual can realistically audit. Often, the code ships anyway.

The gap between fast AI output and slower human verification is at the heart of a new paper by MIT Sloan School of Management research scientist Christian Catalini and his co-authors, Xiang Hui of Washington University and Jane Wu, SM ’18, PhD ’22, of UCLA.

In “Some Simple Economics of AGI,” the researchers lay out an economic model of the transition toward artificial general intelligence — AI systems that can operate with broad autonomy across many tasks. The researchers focus on the problem of measurability: whether the outputs of those systems can be reliably checked.

As AI systems are becoming more capable, it’s getting harder to verify everything they produce. This will put a cap on how fully the benefits of AGI can be realized in the economy: AI makes it cheap to produce work, but not to judge whether that work is any good.

Verification will become a core part of seeing value from AI
AI has largely been considered as a substitute for labor, with the assumption that cheaper output can translate directly into value. But what will distinguish firms is not so much their ability to deploy AI as their ability to stand behind what it produces, the researchers write.

The gap between what AI systems can produce and what can be properly checked is widening. On the SWE-bench AI performance benchmark, the accuracy of AI coding tools rose from 4.4% to 71.7% in a year, and the length of tasks that systems can complete is doubling over short periods, according to the researchers. But there is “scarce capacity” for human verification, as bandwidth continues to be constrained by time and experience, the researchers write.

Verifying AI outputs is no longer just a compliance function but a core part of how value is created from AI, the researchers write. That puts a premium on records of how systems behave — especially where they fail — and on taking responsibility when things go wrong.

Catalini describes this as a shift from software as a service to what he calls “liability as a service.” “The companies that understand the risks and can underwrite them will be the ones that profit,” he said.

The risks of using AI to verify AI and skipping verification altogether
So far, AI adoption has clustered in areas where outputs can be checked quickly: summarizing text, generating images, writing code. But as systems take on longer, higher-risk tasks — and as AI agents act autonomously — checking whether they were done correctly will become more difficult and often take more time, increasing the risk of misplaced trust.

One response from companies is to use AI to check AI, which the researchers call a “tempting shortcut.” But where both systems share the same assumptions, they can reinforce the same errors, creating what Catalini described as a false sense of confidence and not a real solution.

For some companies, competitive pressure and the gap between fast, cheap production and slower verification creates the motivation to deploy systems before they are fully checked by humans, according to the researchers, allowing risks to build unnoticed until they are harder to contain.

But when systems are pushed into use before they are fully verified, the consequences can be severe. Catalini pointed to episodes such as the 2010 flash crash in financial markets, where complex automated systems failed in ways that were not fully understood at the time. “If we do not invest in verification, we’re accumulating hidden risk,” Catalini said. “It is technical debt accumulating behind the scenes, and, at some point, it’ll come due.”

In effect, the economy becomes “hollow,” the researchers write: Output surges, but the quality and utility of the output don’t keep pace. The researchers describe this as a “Trojan horse” problem, where unverified output leaks into the economy and is treated as if it were reliable.

A new paper explores how seeing economic value from artificial intelligence hinges on closing the gap between what AI can do and how humans can verify its outputs.

The Baruch Group

06/23/2026

06/23/2026

06/22/2026

06/22/2026

06/22/2026

06/22/2026

06/21/2026

Address

Telephone

Website

Alerts

Contact The Organization

Shortcuts

Share