FeedExploreAsk AIAlertsSavedProfile

Categories

AICybersecurityInfrastructureDatabaseTech Updates

Tech news that matters.

FeedExploreAskAlertsSavedProfile
Back to feed
AI·High

Why Over Half of Enterprise AI Projects Are Failing

A group of IT professionals discusses a data architecture diagram displayed on a large monitor in an office.

TL;DR: Over half (54%) of companies are abandoning their AI projects, a new report finds. The main culprit isn't the AI model, but rather a poor data foundation that leads to unclear business value and wasted investment.

By Neeraj Dhiman·2h ago·3 min read·updated 21m ago
Source

Key facts

Category
AI
Impact
High
Published
2h ago
Source
TechRadar

Full summary

Over half of enterprises have abandoned their AI projects. The problem isn't the AI—it's the data infrastructure behind it.

A significant number of companies are pulling the plug on their artificial intelligence initiatives before they can deliver results. According to a new report from SnapLogic, cited by TechRadar, 54% of enterprises have abandoned their agentic AI projects. The primary reason cited is a failure to demonstrate clear business value or a positive return on investment (ROI). This is happening despite a general high level of trust in AI outputs and the fact that most organizations already have formal AI policies in place. The report also found that this isn't stopping companies from spending heavily, with two-thirds admitting to moderate or significant overspending on AI, even when budgets are established. The data points to a major disconnect between AI investment and tangible business outcomes, revealing a critical flaw in how many companies are approaching implementation.

This high failure rate isn't because the AI models themselves are flawed. Instead, the report identifies the root cause as a weak data foundation. Agentic AI systems, which are designed to perform tasks autonomously, are entirely dependent on the quality and accessibility of the data they use to make decisions. If a company's data is siloed in different departments, inconsistent, poorly structured, or of low quality, the AI cannot function effectively. It’s a classic “garbage in, garbage out” problem, but at a much larger and more expensive scale. Without a clean, integrated, and reliable data pipeline, the AI is essentially operating with blinders on, unable to generate the insights or perform the actions that would create business value. This leads to pilot projects that stall and full-scale deployments that never get off the ground, ultimately resulting in abandonment.

The findings reflect a broader trend in the technology industry as it moves past the initial hype cycle of generative AI. For the last couple of years, many organizations felt immense pressure to adopt AI to avoid falling behind competitors. This led to a rush to experiment with large language models and other AI tools, often without a clear strategy or the necessary prerequisite work on their data infrastructure. This mirrors previous technology waves, such as the “Big Data” era, where companies collected massive datasets without a plan for how to use them effectively. The SnapLogic report suggests the market is now entering a period of reckoning. The focus is shifting from simply having an AI project to ensuring that AI projects deliver measurable ROI, forcing a renewed emphasis on the unglamorous but essential work of data management.

For CTOs, IT leaders, and founders, the practical takeaway is clear: a successful AI strategy must begin with a robust data strategy. Before investing heavily in complex AI models or ambitious agentic systems, organizations must first assess and improve their data maturity. This involves investing in tools and processes for data integration, governance, and quality control to create a single source of truth that AI systems can reliably draw from. Companies should consider starting with smaller, more focused projects aimed at cleaning up and connecting disparate data sources. Building this solid foundation will not only de-risk future AI investments but also unlock more immediate value from existing data. In the long run, the competitive advantage in AI will belong not just to the companies with the most advanced models, but to those with the best-connected and most reliable data.

Why it matters

This highlights a critical bottleneck for AI implementation that directly impacts engineering and data teams. The failure isn't in the algorithm or model tuning, but in the foundational data architecture. It shifts the focus from model-centric development to the less glamorous but essential work of data integration and governance.

Business impact

Widespread AI project abandonment represents a significant waste of capital and resources. Companies are overspending on AI initiatives that fail to deliver ROI, creating a competitive disadvantage. This report signals that future AI success will depend on strategic investments in data infrastructure, not just expensive models.

Tags

#AI#enterprise ai#it strategy#roi#data infrastructure

Related on Notifire

  • ResearchAI fact-checking for generated content
  • Researchllms.txt
  • ResearchKubernetes security
  • ResearchSoftware supply-chain security

✦ Notifire newsletter

Get more AI intelligence

Join engineers getting Notifire’s verified tech briefings — short, sourced, and free. No spam, unsubscribe anytime.

The day's most important tech briefings. No spam, unsubscribe anytime.

Related stories

Primary source: TechRadar

Tech intelligence for engineering teams

Short, verified briefings on AI, cybersecurity, infrastructure, and data — with the analysis and action steps that matter. Every briefing is sourced, fact-checked, and bylined to a named editor.

[email protected]Story tips & corrections welcomeHow we report →

The Notifire briefing

Verified tech intelligence in your inbox — AI, security, infra, and data.

The day's most important tech briefings. No spam, unsubscribe anytime.

Sections

  • AI
  • Cybersecurity
  • Infrastructure
  • Database
  • Tech Updates
  • Web3 & Chains

Newsroom

  • About Notifire
  • Editorial team
  • Editorial standards
  • Methodology
  • AI disclosure
  • Corrections

Resources

  • Explore
  • Research hubs
  • Comparisons
  • Tech glossary
  • FAQ
  • Alerts & watchlists

Follow

  • RSS feed
© 2026 NotifirePrivacyTermsCorrections
An independent, AI-assisted publication. Built at </Alpheric>
IntelligenceLive panel
Live

Top trending

Last 24h

    Popular tags

    Add to watchlist

    +OpenAI+Claude+PostgreSQL+Kubernetes+Cloudflare+AWS+CVE Critical

    Notifire score

    0–100 priority signal — combines impact, freshness, trending velocity, and source credibility.

  1. Atom feed
  2. LinkedIn
  3. X / Twitter
  4. Facebook
  5. Instagram
  6. YouTube