5 Big Myths of AI and Machine Learning Debunked

Since then, public interest in AI and ML has waxed
and waned. But the release of OpenAI’s ChatGPT
in late 2022, and its competitors that followed,
brought generative AI into the mainstream.
Now it’s more powerful and easier to use than ever before, with use
cases that span across both consumer and enterprise. Generative AI can
plan your next vacation, write poetry in the style of William Shatner or
polish the speech you need to write for your best friend’s wedding. It can
also write policies for a cyber insurance application, generate code or
explain the meaning of a high error in your API service.
Over the next decade, relying on AI in business will be as quotidian
as flipping on a lightswitch. Gartner®’s Hype Cycle™ for Emerging
Technologies 20231 claims that generative AI will reach the peak of
inflated expectations in 2023. However, AI’s long-term potential is
underestimated, says Hao Yang, Splunk’s VP of AI. Generative AI alone
represents an annual $2.6 to $4.4 trillion in opportunity across 63 use
cases, according to a McKinsey study. Spending will only continue to
grow: IDC predicts that worldwide spending on AI will surpass $300
billion in 2026.
Now the ROI of those investments — a historically painful metric to
quantify for AI projects — is starting to crystallize as organizations
expand to more sophisticated use cases. In a PwC survey, 72% of AImature
organizations surveyed (and 59% of all other respondents) are
confident in their abilities to assess the ROI of their current AI initiatives,
with the ability to capture both hard and soft returns and costs.
Despite the promise of AI, an overall mistrust lingers. Fifty-two percent
of organizations say that risk factors are a critical consideration when
evaluating new AI use cases, according to Gartner.2 The old adage “you
can’t protect what you don’t know” rings true for AI and ML. Generative
AI in particular raises deep-seated data privacy and security concerns.
These AI anxieties are only natural as organizations navigate this new
era of technology. The Biden Administration’s executive order on AI
provides a few answers, but substantive regulatory changes are likely
still a ways out.
In the meantime, the AI train has left the station, with positive outcomes
that are too difficult to ignore. Usage is already widespread, with 55%
of respondents in the McKinsey study reporting that their organizations
have adopted AI. And organizations are realizing the value of this
adoption through improvements in productivity, decision-making,
customer experience, innovation and beyond.
To be sure, some of AI’s most far-reaching concepts (computers that
can replicate the human brain entirely, fully autonomous robots and
programs that design, code and upgrade themselves) are years away
from reality; they’re still moonshots that represent the eventual apex
of AI’s capabilities. But considering that AI tools can already win at
Jeopardy!, are able to detect breast cancer and are logging tens of
thousands of miles behind the wheel of self-driving vehicles every
day, the prospect of even those moonshot concepts really doesn’t
seem so far-fetched.
In other words, now is the time to learn about AI and ML. Before you can
develop a thoughtful strategy that considers the risks and benefits, it’s
important to clarify these common misconceptions