The B2B Technology Story Desk: Five stories brands could own in November 2026

Why this story matters now

Enterprises are setting their 2027 technology budgets, and AI investment is moving into a more demanding phase.

The question is no longer whether companies are experimenting with AI. It is whether those experiments are producing enough value to survive financial scrutiny.

A Gartner survey published in September found that only 22% of organizations had successfully scaled AI across multiple business units or adopted an AI-first approach. Yet 85% of functional leaders planned to increase AI spending during 2026.

The same research found that 11% of organizations were entirely unaware of what their function had spent on AI during 2025.

That gap between enthusiasm, financial visibility and successful deployment creates a strong November story.

The questions journalists are likely to ask

• How many AI pilots will be quietly cancelled during 2027 budgeting?
• Do CFOs know what the business is spending on AI?
• Which projects are delivering measurable returns?
• Are companies measuring time saved, costs reduced or revenue generated?
• How much of the true cost sits outside the original software contract?
• Are businesses paying for overlapping AI tools across different departments?
• Which popular use cases are failing to produce meaningful value?
• What happens to AI projects that cannot demonstrate a financial return?

The opportunity for B2B technology companies

This is relevant to companies working in:

• Enterprise AI
• Technology consulting
• Cloud infrastructure
• FinOps
• Data management
• Automation
• Systems integration
• IT asset management
• Technology procurement

The weakest contribution would be another prediction that AI investment will grow.

The strongest contribution would show where the money is going, what businesses are measuring and which projects are being stopped.

A campaign we would consider

Proposed headline:

One in three enterprise AI projects will lose funding in the 2027 budget cycle

Survey 300 CIOs, CFOs, technology leaders and functional buyers about the AI projects currently under review.

Ask:

• How many AI tools or projects the organization funded during 2026
• How many reached production
• How many are used regularly by employees
• Whether each project has a named commercial outcome
• Whether the organization knows its total AI expenditure
• Which costs were excluded from the original business case
• How many projects will receive more, less or no funding in 2027
• Who has authority to discontinue an underperforming project
• Whether productivity gains have been converted into financial value

The research could create several media stories:

• The percentage of AI pilots being cancelled
• The hidden cost of enterprise AI
• Which departments are producing the strongest returns
• The AI applications most likely to lose funding
• The difference between high-performing and low-performing adopters

A stronger proprietary-data version

A FinOps, cloud or enterprise technology company could analyze anonymized customer data to show:

• Growth in AI-related cloud expenditure
• Underused AI infrastructure
• Duplicate services
• Cost per production deployment
• Changes in model or infrastructure usage
• The gap between provisioned and consumed capacity

This would give journalists behavioral evidence rather than executive opinion.

Best suited to: FinOps platforms, enterprise AI providers, cloud companies, technology consultancies, IT asset-management platforms and data infrastructure businesses.

Ideal activation window: October 26 through November 20, while 2027 budgets are still being debated.

Why this story matters now

The European Commission began enforcing significant elements of the AI Act on August 2, 2026.

New transparency requirements mean that certain interactive AI systems must inform people when they are dealing with AI. AI-generated or manipulated content must be identifiable, and deepfakes must be labelled.

The Commission has also published guidelines explaining the responsibilities of providers and deployers.

The immediate compliance deadline has passed. November gives journalists an opportunity to ask whether technology providers have made meaningful changes or simply added another sentence to their terms and conditions.

The questions journalists are likely to ask

• Can users tell when they are interacting with AI?
• What counts as a sufficient disclosure?
• Are companies marking AI-generated content in a machine-readable way?
• Where should disclosure appear inside a product experience?
• Are businesses treating transparency as compliance or as a trust feature?
• Do buyers expect technology vendors to explain which models they use?
• How much information should a company disclose about automated decisions?
• Are transparency requirements creating inconsistent user experiences?

The opportunity for B2B technology companies

The most interesting story is not a summary of the AI Act.

It is whether greater transparency changes how people behave.

A technology company could credibly investigate whether customers are:

• More likely to trust a system that clearly identifies AI involvement
• Less likely to complete a task after an AI disclosure
• More willing to share information with a human than a machine
• Confused by the difference between AI-assisted and fully automated services
• More likely to challenge an automated recommendation
• Interested in knowing which model powers a product

A campaign we would consider

Proposed headline:

Most enterprise software users cannot tell when an AI agent is acting on their behalf

Run a study involving 1,000 business-software users.

Show participants a series of realistic product interactions involving:

• A conventional rules-based workflow
• An AI assistant
• An autonomous agent
• A human supported by AI
• AI-generated content reviewed by a person

Ask participants to identify:

• Whether AI was involved
• What they believed the AI was allowed to do
• Whether a person had reviewed the output
• Who they believed was responsible for an error
• What information they expected the provider to disclose

The results would reveal whether the market’s language around copilots, assistants and agents is understood by the people using these products.

An alternative campaign

Proposed headline:

Clear AI disclosure increases trust, but reduces willingness to delegate high-stakes decisions

Test alternative disclosure formats and measure how they affect:

• Trust
• Task completion
• Willingness to share data
• Willingness to accept an automated recommendation
• Expectations of human oversight

This gives a technology provider a defensible position on responsible product design without resorting to vague claims about “ethical AI.”

Best suited to: Enterprise AI companies, workflow platforms, customer-experience technology providers, legal technology companies, HR technology providers and digital consultancies.

Ideal activation window: Early November, with commentary prepared before Web Summit.

Why this story matters now

In September, the European Commission proposed a common rating system for data centres.

The proposed scheme is intended to increase transparency around energy and water consumption and recognize contributions to the wider energy system, including waste-heat reuse, clean-energy generation and flexible demand.

It would cover individual data centres with capacity above 500 kW. The proposal entered a two-month scrutiny period, placing the issue directly into the November news cycle.

The Commission says Europe wants to triple its data-centre capacity over the next five to seven years. It also says data centres could consume more than 3% of total EU electricity demand by 2030.

The collision between AI growth, infrastructure demand and constrained energy systems will be one of the most important enterprise-technology stories of the next several years.

The questions journalists are likely to ask

• Should data centres receive a public sustainability rating?
• Will technology buyers use the rating when choosing cloud and infrastructure providers?
• Can AI growth continue without placing unacceptable pressure on electricity grids?
• Is water consumption being measured consistently?
• Should data-centre developments be required to reuse waste heat?
• Who pays for the grid upgrades required by new AI infrastructure?
• Can workloads move according to the availability of lower-carbon power?
• Are enterprise buyers willing to trade performance or latency for lower environmental impact?
• Will smaller facilities struggle with the cost of measurement and disclosure?

The opportunity for B2B technology companies

This story extends beyond data-centre operators.

It is relevant to:

• Cloud providers
• Infrastructure companies
• Energy-management platforms
• Cooling technology companies
• Observability platforms
• Workload-optimization businesses
• Enterprise AI companies
• Sustainability technology providers
• Network operators
• Hardware manufacturers

Companies should avoid making broad claims about “green AI” without operational evidence.

Useful contributions would include:

• Real energy consumption by workload
• The effect of model choice on infrastructure requirements
• Water use under different cooling approaches
• How much computing capacity is sitting unused
• Whether workloads can be shifted without affecting customers
• Barriers to waste-heat reuse
• What enterprise technology buyers currently ask about sustainability

A campaign we would consider

Proposed headline:

Enterprise buyers say data-centre sustainability matters, but fewer than one in five include it in procurement decisions

Survey CIOs, procurement leaders and sustainability executives on:

• Whether energy and water use affect infrastructure selection
• What information providers currently disclose
• Whether buyers understand common efficiency metrics
• Whether sustainability requirements appear in contracts
• Whether buyers would accept variable workload timing
• Whether they would pay more for demonstrably lower-impact infrastructure
• Who owns the decision when cost, performance and sustainability conflict

A proprietary-data alternative

A cloud, optimization or infrastructure company could analyze real workloads to calculate:

• Energy wasted through overprovisioning
• Workloads that could move to off-peak periods
• Differences between inference models
• Idle computing capacity
• Potential waste-heat recovery
• Regional differences in energy intensity

Proposed headline:

A quarter of enterprise AI infrastructure is provisioned but rarely used

The exact headline would depend on the data, but the principle is to quantify the infrastructure inefficiency created by rapid AI adoption.

Best suited to: Cloud platforms, infrastructure providers, data-centre technology companies, FinOps platforms, energy-management businesses and enterprise AI infrastructure companies.

Ideal activation window: November 9 to November 27, particularly if the EU scrutiny period generates policy activity.

Why this story matters now

Web Summit takes place in Lisbon from November 9 to 12.

It will bring technology companies, investors, founders and journalists into the same venue. It will also create an intense competition for attention.

Many companies will wait until the event to announce:

• A product
• A funding round
• A partnership
• A new market
• An AI feature
• A rebrand
• A customer
• A research report

The problem is that journalists will receive hundreds of broadly similar pitches containing the same claims: faster, smarter, transformative, AI-powered and industry-first.

The opportunity is not merely to attend Web Summit. It is to give journalists a story they can use while everyone else is offering announcements.

The questions journalists are likely to ask

• Which technology claims are supported by evidence?
• Are startups solving important problems or adding AI to existing products?
• What has changed since the previous year?
• Which companies have real adoption rather than pilot customers?
• What are investors funding now?
• Are European technology companies becoming more commercially disciplined?
• Which AI companies can explain their route to profitability?
• What do enterprise buyers actually want from the technologies being launched?

The opportunity for B2B technology companies

A successful Web Summit media strategy should begin before the event.

Journalists will have limited time and overloaded schedules. A company needs at least one of the following:

• Original data
• A credible contrarian position
• A meaningful customer result
• A visually demonstrable product
• A significant business announcement
• Access to an unusually informed spokesperson
• A story that connects to a wider market shift

A campaign we would consider

Proposed headline:

Enterprise buyers say most technology launches fail to explain the business problem they solve

Before Web Summit, survey 300 enterprise technology buyers about the language used by vendors.

Test:

• Which claims buyers trust least
• Which product terms they do not understand
• Whether “AI-powered” increases or reduces interest
• What proof buyers expect from a new technology company
• Whether buyers prefer product capabilities or customer outcomes
• What makes them agree to a meeting
• How many tools they are actively trying to remove from their stack

Release the findings several days before the event, then use the company’s presence at Web Summit to discuss what enterprise buyers actually want.

A more provocative alternative

Analyze the websites or launch materials of 500 B2B technology companies exhibiting at Web Summit.

Measure the prevalence of terms such as:

• AI-powered
• Transformative
• Revolutionary
• Seamless
• Next-generation
• Intelligent
• Industry-leading
• All-in-one

Then compare that language with whether the company provides:

• A quantified customer result
• A clear use case
• Transparent pricing
• Named customers
• A clear description of the buyer
• Evidence of integration
• A measurable business outcome

Proposed headline:

“AI-powered” appears more often than customer evidence across Web Summit exhibitors

The campaign would need to be conducted fairly and transparently, but it could give a company a distinctive voice during an extremely crowded media week.

Best suited to: Enterprise technology platforms, procurement technology companies, technology consultancies, product-experience platforms and companies attending or exhibiting at Web Summit.

Ideal activation window: Teaser findings from November 2, full release between November 5 and 9, rapid commentary throughout November 9 to 12.

Why this story matters now

The enterprise technology conversation is moving from tools that generate content to systems that can initiate and complete actions.

These systems can:

• Update records
• Contact customers
• Approve routine requests
• Change configurations
• Purchase services
• Move data between systems
• Recommend or initiate decisions
• Trigger other automated workflows

The technical capability is developing faster than many organizations’ governance structures.

IBM’s 2026 workforce research found that only 26% of organizations clearly distinguish between human-led, AI-assisted and AI-executed activities. It also found that 36% of executives see unclear accountability as a complication in AI deployment.

That creates an important B2B technology question:

When an AI agent makes a commercially significant mistake, who owns the decision?

The questions journalists are likely to ask

• What decisions should an AI agent never make independently?
• Who is accountable when an agent acts incorrectly?
• Can organizations reconstruct why an automated action occurred?
• Do employees know when they can override an AI system?
• Are agents receiving excessive access to company data and systems?
• How should companies approve new agentic workflows?
• Is human oversight meaningful or merely nominal?
• Are businesses monitoring agent-to-agent activity?
• What happens when an agent follows its instructions but produces a damaging outcome?

The opportunity for B2B technology companies

This is a valuable story for businesses working in:

• Enterprise AI
• Governance
• Observability
• Identity and access management
• Workflow automation
• Data infrastructure
• Compliance
• Enterprise architecture
• Technology consulting

Avoid another general warning that “governance is important.”

The strongest contribution would reveal what organizations are actually allowing agents to do and whether responsibility has been assigned.

A campaign we would consider

Proposed headline:

Half of businesses using AI agents cannot identify who is responsible when an automated decision goes wrong

Survey CIOs, technology leaders, risk executives and operational leaders at companies deploying AI agents.

Ask:

• Which actions agents are permitted to complete
• Which systems agents can access
• Whether every agent has a named human owner
• Whether actions are logged and reconstructable
• Whether financial limits exist
• When human approval is required
• Whether the organization has tested an agent failure
• Who is responsible for customer remediation
• Whether agents can communicate with other autonomous systems
• Whether access is removed when a project ends

A practical demonstration

A technology company could also build an interactive “agent responsibility test” for business leaders.

Participants would be shown scenarios such as:

• An agent issues an incorrect customer refund
• An agent changes a pricing rule
• An agent sends confidential information to a supplier
• An agent rejects a job candidate
• An agent purchases additional cloud capacity
• An agent deletes a record it considers redundant

For each scenario, participants select who they believe is responsible.

The disagreement between technology, legal, operational and board respondents could become the story.

Best suited to: AI governance platforms, observability companies, identity providers, automation businesses, compliance technology providers and enterprise consultancies.

Ideal activation window: Mid to late November, after Web Summit has amplified interest in agentic technology.