Persistent Technology Challenges Companies Still Need To Solve

gettyTechnology leaders often have no trouble identifying the problems holding their organizations back. The harder part is turning conversations about those challenges into sustained action, particularly when meaningful change requires rethinking familiar priorities, practices or assumptions.

Recognizing a problem is only the starting point; real progress depends on defining what improvement should look like and following through. Below, members of Forbes Technology Council share issues tech leaders frequently discuss but struggle to address and explain what real progress could look like.

Knowledge silos are the one thing everyone talks about, but nobody fixes. Critical systems live in one person’s head instead of being documented, and it never gets prioritized because docs don’t hit deadlines. Real fix: Build coverage into how work gets assigned and actually track it like uptime. Bottom line: It’s an incentive problem, not a technical one. - Tiffany Saylor, FB Society

One issue is value creation. Organizations spend large amounts on technology and measure success by whether the project launched on time. After 1,500 implementations, I can tell you that going live is where the real work begins. Real progress means tracking whether the tool creates value for the organization. - Tal Frankfurt, Cloud for Good

Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?

One challenge that doesn’t get enough attention is that everyone defines AI differently. Without a shared vocabulary, it’s difficult for business leaders, technology providers and customers to align on expectations, outcomes and value. Broad adoption requires a common language for AI and establishing frameworks, best practices and standards that make AI easier to understand, implement and scale. - Nick Heddy, Pax8

Everyone’s talking about AI, but not enough is being said about the disconnected data undermining it. Small and mid-size businesses run up to 25 different apps to get a full view of their data, and that sprawl caps their ability to get real insight. This strains finance teams and forces them to focus on manual workarounds instead of the strategic work AI should free them up for. Real progress means giving these businesses clean, connected data by default. - Ashley Still, Intuit, Inc.

The gap between vision and infrastructure: Everyone wants to talk about what the future will look like, but not enough attention goes to the underlying data, interoperability and operational backbone required to make it work reliably at scale. Real progress doesn’t look like a better promise or a better demo; it looks like foundations strong enough to make the vision practical. - Michael Harrell, TomTom

Many tech leaders talk about AI transformation within their teams and company but approach it as a software rollout. They provide tools, track adoption and expect productivity to improve. Real progress means reconsidering roles, workflows, approvals, team structures and decision rights. The opportunity is not simply to help people perform their existing tasks faster; it is to eliminate unnecessary tasks and handoffs altogether. - Roby Baruch, Tipalti

Tech leaders continue to talk about AI return on investment but lack a consistent measurement of whether their investments create value. One of the biggest opportunities to capture AI ROI is through human judgment, or the revisions applied to AI-generated outputs. A judgment system documents how first drafts evolve into final deliverables. Organizations will differentiate themselves by applying better judgment to outputs and turning it into business value. - Samir Dutta, Farsight AI

One issue is data quality. Leaders celebrate new tools and faster systems but rarely ask whether the underlying data is actually trustworthy. Real progress means validating information at the source, before it corrupts every downstream decision. In healthcare, for example, AI can now capture clinical notes instantly, but too often it’s built on the same unvalidated data it always was. We’ve sped up the conveyor belt without fixing what’s on it. - David Lareau, Medicomp Systems

Tech leaders talk about allyship and showing up for others but often fail to advocate for them when they’re not present. For example, supporting a female colleague at a presentation is not enough. Advocating on her behalf when she’s not in the room opens up new opportunities for her and gives her the ability to demonstrate her capabilities. - Diana Cano, Cambium Learning Group

Many enterprises are racing to automate AI decisions on unsupported foundations. AI is only as trustworthy as its governance. Real progress begins when trust becomes core infrastructure through validating data, preserving transparency and keeping humans accountable. The future isn’t about replacing human judgment but amplifying it with explainable, verifiable intelligence worthy of trust. - Kenneth Coats, KENTECH

Application rationalization is a priority almost everyone acknowledges but too few operationalize. A CHIME survey found that 76% of CIOs say it is critical to their application portfolio strategy, yet 80% have not fully implemented an ongoing program. Real progress means giving an application undertaker an enterprise mandate to retire legacy apps and remove technical debt. - Jason Rose, Clearsense

The biggest challenge tech leaders underestimate is the human side of AI adoption. Many organizations focus on selecting models, building capabilities and launching pilots but overlook the behavioral shift required to change how people work. Progress comes when leaders pair AI implementation with change management, helping employees understand its value, build confidence and integrate it into everyday workflows. - Nitesh Mirchandani, Mindsprint

Technical debt from AI systems shipped fast is talked about constantly but rarely addressed—until it breaks something. Every panel raises responsible AI and then moves on, because fixing debt doesn’t show up on a quarterly scorecard. Real progress means treating model debt like a liability—tracked on a ledger, owned by a named person and paid down on a schedule, not “when we have time.” - Ankur Pal, Aplazo

Tech leaders often discuss but still fail to address how to keep humans in control of AI. They focus on making AI more capable but not on defining where human judgment must remain final. Real progress means designing AI to support decisions, not replace accountability. Humans must remain responsible for high-impact choices, with clear oversight, transparency and the ability to override AI when needed. - John Davagian, L2L

AI’s real gap isn’t intelligence; it’s trustworthy governance. Powerful models grab headlines, but in healthcare, finance and infrastructure, capability alone isn’t enough. Progress means transparent, auditable and corrigible systems aligned with human goals. The future belongs not to the most powerful models but to the most trusted ones. - Guy Leitersdorf, Longevity AI

One of the real fears I see every day is among tech team members who are wondering how their careers will progress, what they need to learn, and how they can remain ahead of the curve and continue to leverage their innovative spirit to execute on company vision through technology. We should not minimize the effects that advancements in technology create across companies whose main product is that technology. - Eglae Recchia, Keyway

One issue leaders struggle with is moving AI from experimentation to measurable impact. Too often, teams spend months perfecting a model before real users see it. Real progress comes from launching early with a defined user group, gathering direct feedback and measuring clear KPIs around adoption, productivity and outcomes. AI becomes meaningful when you show it changes how people work, not simply that the technology exists. - Siamak Baharloo, Labviva

One issue is addressing end-to-end process gaps that result in friction in the ways technology supports the business. Business partners often have many workarounds due to system limitations. However, these gaps in the system rarely get prioritized in future enhancements due to competing priorities. Focusing on addressing these gaps with forward-deployed engineers who work alongside the business would provide meaningful value. - Marc Kermisch, Protolabs

AI adoption is outpacing security oversight. Boards push efficiency, CEOs drive rollout and CISOs, often without an equal seat at the table, absorb the resulting risk, especially from AI-generated code entering enterprise codebases unchecked. Skilled human oversight is far from obsolete. AI software governance, tooling traceability and role-based security upskilling must be boardroom priorities, considered and implemented at the same pace as adoption. - Pieter Danhieux, Secure Code Warrior

Tech leaders talk about becoming “data-driven,” yet most organizations collect far more data than they turn into action. Another dashboard is not progress. Real progress means contextual intelligence: understanding what’s happening in the moment and adapting accordingly. In physical environments, this can improve service, personalize engagement, reduce friction and create revenue. Success is better experiences and measurable outcomes, not more technology. - Imre Szenttornyay, CieloVision