Hewlett-Packard is in trouble. Smart phones and tablets threaten to trivialize its status as the world’s largest PC maker; cloud computing is destabilizing its server business; printers are no longer seen as high tech. HP’s stock price plunged in 2011 as the company groped for a strategic direction.
Part of new CEO Meg Whitman’s plan to revitalize HP is to lean more heavily on ideas from the 500 researchers at HP Labs, which is headquartered in Palo Alto, California. They are explicitly tasked with creating disruptive technologies—even if these challenge HP’s current business. (For more on disruptive innovations, see Business Impact.)
One of Whitman’s first steps was to make the labs’ director, electrical engineer Prith Banerjee, report directly to her. Banerjee told Technology Review IT editor Tom Simonite why disruptive technology is a helpful internal force at HP, not only an unwelcome external one. Among the labs’ recent achievements: persuading the company to try low-cost server chips and pioneering a novel memory component known as the memristor.
TR: What’s the biggest advance that HP Labs has made during your tenure?
Banerjee: We developed a Richter accelerometer 1,000 times more sensitive than existing sensors. We are working with Shell on using that for energy exploration, combining HP sensors, networking, servers, and software to provide a better picture of existing and alternative energy resources. It will allow Shell to perform more targeted discovery and reduce environmental impact.
Most people associate disruptive technology with startup companies. Why should we believe big companies can innovate?
Startups take research from academia and innovate around how to bring it to market—the execution. The majority of them do very little R&D. A startup could not invent a technology like the memristor: it took us five to 10 years to come up with the right combination of materials and everything else. We are doing fundamental innovation and bringing that to market within the next three years.
What is an example of an idea from the labs that HP is adopting even though it disrupts the existing business?
Project Moonshot, where we launched servers based around low-powered chips with an ARM or Intel Atom architecture like those in mobile devices. Our server business is a very lucrative cash cow; we want people to buy these hunks of metal with large profit margins. But my researchers could see that the future is not heavy number-crunching—it is the tweeting, real-time-updating Zyngas of the world. The architecture we need for that is not a server with a high-performance Intel processor—it’s a lot of low-powered processors working together. Eventually the head of our server business said, “We’re going to do this.”
Isn’t it difficult to get leaders in the business units to listen to ideas that challenge their current products?
There are times it’s not enough to explain something or show a research prototype. We have a program called Demonstrators to build a product from scratch that really convinces people. For example, we had been researching photonic [networking] interconnects—that use light, not electricity—for several years. But the businesses said it was expensive and couldn’t be built. We took a very high-end router that HP makes and changed the [internals] into photonics. When we demoed that, everyone said, “Wow, this is a great innovation.”
HP has struggled with smart phones and tablets. How is HP Labs helping the company respond?
We have a mobile team in the lab that works hand in hand with the business; we are passing them the tools to adapt. This is a very fast-moving market, so we can’t talk about it in detail. We also are researching areas that HP is not currently working in, and that goes for our mobile team, too.
HP Labs has sites in the U.K., Israel, Russia, Singapore, India, and China. Does that help you find disruptive ideas?
If you are only U.S.-centric, you are missing a lot of opportunities. Let’s say you’re trying to create a tablet device for India. You would think of the iPad, but out of India’s one billion people, only 50 million access the Internet—but 500 million have access to simple phones. Our researchers in India built a hardware device called the Vayu and a whole cloud system to deliver the Web via SMS, called Siteonmobiles. Our researchers in Palo Alto would never have thought of that.
After you joined HP Labs, you consolidated research into fewer areas. Why?
We were working on far too many cool things. We had 150-ish projects, each involving one or two researchers. We felt that we needed to channel the creative minds of these people. If every project would bring together two computer scientists, one chemical engineer, a social scientist, and a physicist, something really cool could happen.
This new data poisoning tool lets artists fight back against generative AI
The tool, called Nightshade, messes up training data in ways that could cause serious damage to image-generating AI models.
The Biggest Questions: What is death?
New neuroscience is challenging our understanding of the dying process—bringing opportunities for the living.
Rogue superintelligence and merging with machines: Inside the mind of OpenAI’s chief scientist
An exclusive conversation with Ilya Sutskever on his fears for the future of AI and why they’ve made him change the focus of his life’s work.
How to fix the internet
If we want online discourse to improve, we need to move beyond the big platforms.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.