Connected, But Less Protected | The Decision Signal

The Observation

For years, companies have invested in technology with the expectation that greater connectivity and automation would make their operations faster, more productive, and ultimately more profitable. Those advantages remain compelling, but this week offered a particularly striking reminder that the same capabilities are becoming available to those looking to exploit them.

A single hacker reportedly used autonomous AI agents to penetrate more than 100 company websites at an estimated computing cost of approximately $50 per successful intrusion. A compromise involving a cloud backup provider interrupted services used by other businesses. A major international grocery retailer continued restoring digital operations after taking systems offline to contain an attack. Meanwhile, AI-assisted research exposed serious vulnerabilities in software underlying much of the internet, prompting organizations to undertake emergency repairs. The financial consequences deserve attention beyond the technology department.

An unavailable application can interrupt revenue. Compromised credentials can expose customers, contracts, and intellectual property. Recovery consumes operating budgets, while additional investment in security competes with other demands for capital. Even precautionary shutdowns introduce a difficult calculation between the cost of remaining operational and the potential consequences of continuing. There is another dimension to this equation. The technologies making cyberattacks less expensive are also helping defenders identify vulnerabilities, strengthen their systems, and respond more effectively. Organizations choosing to limit AI deployment may reduce some immediate exposure, but they also risk passing up tools that could improve productivity and reduce the cost of protecting their businesses.

The challenge for executives is deciding how to capture those benefits without expanding financial and operational exposure beyond what the organization can responsibly manage. As companies prepare their 2027 operating plans, intelligence is becoming both an investment opportunity and an increasingly important consideration in the economics of business continuity.

Friction / Traction Index

Select any condition to compare its historical trend with the overall index.

Rolling 12 Weeks

August 3 to October 23, 2026

The Friction / Traction Index measures four business conditions: Demand, Capital, Cost, and Execution. Each is scored from −2 (strong friction) to +2 (strong traction). The Overall Index is the equally weighted average of these four conditions.

The Evidence

Demand

Firming · Uneven activity

Business activity is improving, though customers remain selective.

The Philadelphia Fed’s October nonmanufacturing survey showed new orders rising from −2.2 to 2.8 and sales/revenues from 6.8 to 14.6. Abbott reported 7% comparable sales growth, while pending home sales fell 7.3%.

Ahold Delhaize’s cyber disruption also exposed the difference between customers willing to buy and businesses able to serve them.

Demand is strengthening in pockets, although performance remains uneven across industries and purchasing channels.

Sources: Philadelphia Fed · Abbott · National Association of Realtors · Ahold Delhaize

Capital

Constrained · Narrowing investment

Investment continues, but fewer firms report expanding commitments.

The Philadelphia Fed’s October nonmanufacturing survey showed the equipment and software capital expenditure index falling from 33.0 to 18.9, and physical plant from 22.9 to 13.0.

Separately, equipment-finance executives expecting increased demand for capital expenditure financing declined from 31.8% to 18.2%.

Investment remains positive in both surveys, but its expected expansion is becoming less widespread. For consultancies, that can mean longer decisions on new engagements even as existing client activity improves.

Sources: Philadelphia Fed · Equipment Leasing & Finance Association

Cost

Deteriorating · Expanding pressure

Input costs and risk exposure are rising together.

Procter & Gamble now expects $1.4 billion in higher after-tax raw material, energy, and transportation costs for fiscal 2027, up from $1 billion.

Marsh reported flat U.S. cyber insurance rates, a 4% global decline, and a 6% increase in U.S. casualty liability rates. Travelers reported a 57% year-over-year increase in the frequency of financial fraud and business-email-compromise cyber claims in the first half of 2026.

Stable insurance premiums do not necessarily mean stable financial exposure. Deductibles, interrupted operations, and uninsured losses can still erode margins.

Sources: Procter & Gamble · Marsh · Insurance Journal

Execution

Improving · Uneven gains

Productivity is improving, while digital dependencies deepen.

Union Pacific increased workforce productivity 5% and improved fuel efficiency 2%. GE Aerospace increased year-to-date engine deliveries by more than 20%, supported by record internal shop-visit output.

AI-assisted research exposed critical PHP vulnerabilities, enabling coordinated repairs. ConnectWise, however, suspended cloud backup and related services after unauthorized access to application identities.

Greater output and better detection are measurable gains. Their value also depends on technology providers and essential systems remaining available. For consultancies, performance includes the reliability of the systems they help clients operate.

Sources: Union Pacific · GE Aerospace · OpenAI · ConnectWise

Five Important Reads

Five perspectives on the changing economics of intelligence, cyber exposure, and the financial choices facing business leaders.

01.Economics of cybercrime

How one hacker used AI agents and a few thousand dollars to break into scores of sites

Reuters, via BNN Bloomberg · October 19, 2026

A researcher documented an AI-assisted hacking campaign involving more than 1,200 attempted intrusions and over 100 successful website compromises. The estimated computing expenditure was approximately $5,000 for the campaign, with limited human intervention.

Why it matters: AI is reducing the effort and expense required to launch attacks at scale. Businesses may face greater exposure even without becoming particularly valuable or prominent targets.

02.Defensive intelligence

What to know about the AI-detected security bug impacting wide swaths of the internet

Associated Press · October 22, 2026

AI-assisted vulnerability research helped expose serious security flaws in PHP, a programming language widely used by websites and applications. Organizations began deploying emergency patches, with some temporarily suspending services to reduce their exposure.

Why it matters: AI is creating new risks while improving the ability to discover existing ones. The economic value of adoption includes losses that better detection and remediation may help organizations avoid.

03.Third-party exposure

ConnectWise breach: Cloud services disruption continues three days after initial intrusion detected

ITPro · October 20, 2026

A compromise involving ConnectWise cloud backup application identities enabled unauthorized access attempts against customer Microsoft 365 environments. The company suspended affected services and began remediation, leaving downstream organizations responsible for investigating potential changes in their own environments.

Why it matters: Outsourcing a technology function does not eliminate the financial consequences of its failure. Executives need to understand how the permissions granted to suppliers could affect continuity, customer information, and recovery obligations.

04.Risk transfer

Global commercial insurance rates fall 5% in Q3 2026

Marsh · October 21, 2026

Marsh found that global commercial insurance rates fell 5% in the third quarter. Cyber pricing declined 4% worldwide but remained flat in the U.S., while U.S. casualty liability pricing increased 6%. Some buyers also obtained broader coverage, higher limits, or lower deductibles.

Why it matters: Lower insurance premiums do not necessarily mean lower financial exposure. The terms governing interrupted operations, third-party failures, and retained losses become increasingly important as businesses deploy more intelligence.

05.Return on intelligence

A saved hour from AI is not yet a cash saving

CFO.com · October 20, 2026

An examination of how finance leaders should distinguish between productivity improvements and financial returns from AI investments. Time saved, capacity created, and actual expenditure reductions are different outcomes, requiring different measures.

Why it matters: Organizations need an economic basis for deciding where to deploy intelligence. Security, efficiency, and resilience may all justify investment, but their benefits must be measured according to the financial outcomes they produce.