Industry executives and experts share their predictions for 2025. Read them in this 17th annual VMblog.com series exclusive.
By Rick Pastore
You’ll see dozens of predictions calling 2025 the year AI goes
mainstream. You’ll recall the same predictions for 2024 and 2023. The reality
is that most corporate GenAI deployments are still in pilot stages. Joining the
continued hype is a new undercurrent of backlash and disappointment that GenAI
is not showing payoff. Meanwhile, fears about AI’s weaponization and
undermining of social trust continue unabated. And it’s all happening at an
unprecedented pace. According to a Gartner survey, the use of AI technologies,
including GenAI, in enterprises increased from 25% in 2019 to over 50% in 2023,
a doubling of adoption within just a few years. Indeed, the speed at which
GenAI tools reached widespread usage – 100 million users in less than a year –
surpasses adoption rates seen in the early stages of cloud computing and
robotic process automation, to name two transformative technologies.
SustainableIT.org’s AI prediction is a bit different. This coming year
will see the start of responsible and sustainable scaling of AI in businesses.
The term responsible AI has caught
on. There are nonprofit institutes and university programs devoted to it.
Generally, responsible AI refers to human safety, inclusion and accessibility.
But SustainableIT’s perspective, developed by the information technology (IT)
executives that founded and subscribe to our non-profit professional
association, adds environmental sustainability to the concept. To us,
responsible AI is the development
and implementation of AI systems at scale in a way that realizes value while
ensuring fairness, accountability, and ethical alignment. Responsible AI meets
regulations, minimizes bias, maximizes accessibility, and protects privacy while
mitigating harms such as discrimination, misuse, and environmental degradation
through emissions.
The coming year will see Generative AI and Large Language Model
applications moving out of pilots and being scaled to business functions, units
and enterprises. But rather than ungoverned proliferation, as often happens
with promising (and much-hyped) new technologies, IT executives will exert
their influence and leverage their well-tested deployment guardrails to ensure
the AI tools scale in a way that unleashes their value but also protects and
supports corporate commitments to net zero emissions targets and stakeholder
accessibility and inclusion; elevates employee skill sets and capabilities;
fulfills compliance requirements for data protection, transparency and safety;
and does all this cost effectively with as little rework or waste as possible.
In their capacity as the de facto governing authority of enterprise
technology, IT leaders within
SutainableIT.org have developed and published these goals in the form of
the following nine principles applied to the deployment and ongoing use of AI
in the enterprise.
1. Risk due diligence: AI
applications are thoroughly analyzed before deployment at scale for their risk
materiality to, and implications for, business operations, policies,
compliance, goals and strategies.
2. Sustainability due diligence: AI
applications are thoroughly analyzed before deployment at scale for current and
long-term implications for environmental, social, and governance commitments,
policies, and regulations.
3. Ethical usage: AI
application deployment and use are monitored for alignment to and compliance
with the organization’s ethical standards and business values (e.g., equity and
inclusion, nondiscrimination, transparency, safety).
4. Data optimization: Data
used in AI applications are appropriate, transparent, secure, privacy
compliant, consensual, and as unbiased as possible, supported by appropriate
data governance.
5. Trustworthy outcomes: Results,
recommendations, and decisions made or informed by AI applications are fair,
reasonable, explainable, accurate, and cause no harm to human health, safety,
or fundamental rights.
6. AI literacy: AI
application deployment coincides with development of users’ understanding of AI
operations, limitations and risks, and the knowledge to apply AI appropriately
and effectively in their roles.
7. Human first: AI deployment
prioritizes enhancement and augmentation of existing jobs/roles (Human + AI),
with upskilling and prioritized redeployment of displaced workers.
8. Inclusive benefits: The
benefits of AI are equitably applied to, accessible, and leveraged by the
broadest range of targeted stakeholders and do not intentionally or
unintentionally exclude disadvantaged groups.
9. Responsible innovation: AI-dependent
innovation goals and outcomes are subject to the same governing criteria for
risk, sustainability, and ethics that are applied to AI usage.
Members of SustainabeIT.org are developing a “runbook” of actions to
implement and operationalize these principles, which they will share with the
global community in 2025. On top of that, they are developing guidelines for
persona-based employee AI literacy programs, governance resources including a
catalog of AI implementation frameworks, and models for data integrity model
and cost-benefit of responsible development and deployment. They will also be
engaging with the UN Global Digital Compact and AI For Good leadership to
provide these resources for incorporation In UN guidelines and global exposure
to the private and public sectors.
IT leaders don’t have all the answers to responsible AI, and many of the
principles for its governance are not fully under their control. However, the
profession has experienced enough innovative technology scenarios to know that
AI is on a potential trajectory to costly implementation misfires, publicly
embarrassing data miscues, costly compliance lapses, and exacerbation of the
digital divide. Add to that the impact of AI’s unprecedented energy demands,
and you have the potential for technology-induced corporate havoc of
unprecedented proportions.
However, as more and more CEOs, national leaders, corporate investors
and public watchdogs realize the AI risk as well as the reward that AI poses,
we feel confident that responsibility will prevail. With IT executives at the
AI strategy table, 2025 will be the year we start seeing the smart, sustainable
scaling of what may still prove to be the most transformative technology tools
in history.
##
ABOUT THE
AUTHOR
Richard Pastore is Research Principal of SustainableIT.org, a
501(c)(6) nonprofit organization founded and led by technology executives to
advance global sustainability through technology leadership. He has over 25
years of experience working with chief information and technology officers to
apply thought leadership and best practices to maximize business value from
information technology. Mr. Pastore has spent the last 15 years designing,
implementing and managing IT and business transformation leadership programs,
including best practices research, seminars, workshops, assessment tools, and
frameworks for Global 1000 companies. He is former editor of CIO magazine and cio.com,
vice president of the CIO Executive Council, and senior director of research at
The Hackett Group.
