Don’t Believe the Hype: AI Not a Magic Bullet in Drug Discovery

Aria Pharmaceuticals
3 min readOct 9, 2019

--

As with other transformative technologies, expectations will be raised, lowered, dashed and restored before AI shows its real value

by: Andrew A. Radin
CEO,
twoXAR

Since co-founding twoXAR five years ago, I have consistently held the same beliefs about AI’s role in drug discovery, and I’ve delivered a consistent message about it. The world’s reaction to my words has evolved, from skepticism to curiosity to full-blown excitement.

Many believe AI is the panacea that will revolutionize the healthcare industry, and dollars are flowing into healthcare AI startups at unprecedented rates as investors make big bets on using the technology to analyze health data, improve efficiency in healthcare delivery, speed drug discovery, and offer new therapies for intractable diseases.

Now, it’s time to prepare for the next phase of the AI journey: the trough of disillusionment.

The trough of disillusionment is the third of the five phases of the “hype cycle,” a concept developed by the research and advisory firm Gartner. This is the phase where the drug discovery community shall declare that AI is total BS. Well, almost.

The Hype Cycle

What sets us up for the trough of disillusionment is the phase that comes right before it, the steady climb up to the peak of inflated expectations. In the case of AI-driven drug discovery (AIDD) companies, the peak is built upon the momentum of early players like twoXAR: promising research emerges that attracts seed financing. That seed financing leads to promising experimental results. Those proof-of-concept studies attract early adopters to investigate what is possible. News of those early collaborations and scientific results fuels the media, and the hype begins.

As activity moves beyond the early adopters, organizations increase their commitments despite an incomplete understanding of the capabilities of AI. New players enter the space, building upon the momentum to power their own business activity. Before you know it, we have hundreds of players all making the claim that they have AI and it’s going to change drug discovery in a fundamental way.

The AIDD space is fully primed to go over the peak and fall into the trough. It will start with negative press. Over-hyped companies will get a haircut on valuation. Purportedly advanced technologies will not deliver to expectations. Prior announcements on big-dollar deals will stumble in fulfilling their promises to transform healthcare. Over-hyped and over-valued companies will dissolve or merge with stronger players, and it will appear as if the whole movement was a bust.

The early signs of the slide are showing. In April 2019, the seemingly powerful IBM Watson announced earlier this year they are pulling back on drug discovery efforts with partner MD Anderson Cancer Center over unfulfilled expectations. And last month, Forbes reported that promising company BenevolentAI is preparing to slash its multi-billion dollar valuation.

But remember: The hype cycle also includes phases of recovery and productivity.

The good news for AIDD companies is that the strongest players will continue to build a steady stream of results, even if they come more slowly and with less fanfare than founders, investors and the media had wanted. This steady stream of empirical evidence will drive us to the slope of enlightenment — where the scientific community will come to a new understanding of what to expect from the emergence of AI. We can expect, in time, to advance to a plateau of productivity where the true value of artificial intelligence is delivered.

I look forward to the next five years of twoXAR’s journey as we make our way through the peaks and valleys of the hype cycle. When we emerge at the other end, in collaboration with the entire AIDD space, we will do what we set out to do: improve the lives of patients in need.

--

--

Aria Pharmaceuticals

Aria Pharmaceuticals is a preclinical-stage pharmaceutical company discovering and developing novel small molecule therapies for complex diseases.