Blogs

Accidents happen, but not all of them are inevitable. Drunk driving is one of the deadliest and most preventable causes of roadway fatalities. In 2022 alone, more than 13,000 people died in alcohol-related vehicular crashes in the United States, accounting for nearly a third of all traffic deaths, according to the National Highway Traffic Safety Administration.Now a group of high school students in North Carolina is taking action with SoberRide, an AI-enabled device they designed to prevent intoxicated people from driving.Breathalyzer-based ignition interlocks are already in use; they require the driver to blow into a device, proving they are sober enough to drive. However, these interlocks are not foolproof because someone other than the driver could breathe into them, trying to outsmart the device.SoberRide uses a combination of cameras, sensors, and machine-learning algorithms to detect signs of alcohol impairment in the driver—such as pupil dilation, bloodshot eyes, and the presence of ethanol used in alcoholic beverages—before allowing a vehicle to be put into drive. “We’ve been training our neural network to classify intoxication, refining the system’s ability to reliably sense whether someone is drunk or sober,” says Swayam Shah, chief executive officer and cofounder of SoberRide. He’s an 11th-grader at Enloe Magnet High School, in Raleigh.The SoberRide team presented its invention at the MIT Undergraduate Research Technology Conference in October, sponsored by groups like the IEEE University Partnership Program and IEEE Women in Engineering.The students also showcased their technology at another IEEE-supported event: the International Conference on Artificial Intelligence, Robotics, and Communication, held in December in Xiamen, China. From tragedy to technology The inspiration for SoberRide came from a tragedy. Shah was in eighth grade when a neighbor was killed in a collision caused by a drunk driver. The loss prompted Shah to research the magnitude of the drunk-driving problem. “We learned that nearly 300,000 people die each year in crashes involving at least one drunk driver,” says Shaurya Mantrala, a senior at Enloe and the […]

Coding assistants like GitHub Copilot and Codeium are already changing software engineering. Based on existing code and an engineer’s prompts, these assistants can suggest new lines or whole chunks of code, serving as a kind of advanced autocomplete. At first glance, the results are fascinating. Coding assistants are already changing the work of some programmers and transforming how coding is taught. However, this is the question we need to answer: Is this kind of generative AI just a glorified help tool, or can it actually bring substantial change to a developer’s workflow? At Advanced Micro Devices (AMD), we design and develop CPUs, GPUs, and other computing chips. But a lot of what we do is developing software to create the low-level software that integrates operating systems and other customer software seamlessly with our own hardware. In fact, about half of AMD engineers are software engineers, which is not uncommon for a company like ours. Naturally, we have a keen interest in understanding the potential of AI for our software-development process. To understand where and how AI can be most helpful, we recently conducted several deep dives into how we develop software. What we found was surprising: The kinds of tasks coding assistants are good at—namely, busting out lines of code—are actually a very small part of the software engineer’s job. Our developers spend the majority of their efforts on a range of tasks that include learning new tools and techniques, triaging problems, debugging those problems, and testing the software. We hope to go beyond individual assistants for each stage and chain them together into an autonomous software-development machine—with a human in the loop, of course. Even for the coding copilots’ bread-and-butter task of writing code, we found that the assistants offered diminishing returns: They were very helpful for junior developers working on basic tasks, but not that helpful for more senior developers who worked on specialized tasks. To use artificial intelligence in a truly transformative way, we concluded, we couldn’t limit ourselves to just copilots. We needed to think more holistically about the […]

Henry Samueli, cofounder of Broadcom and the 2025 recipient of the IEEE Medal of Honor, has this advice for engineering students and recent graduates just starting on their careers:“Don’t do engineering for the money. Do it because it can have an impact, because you enjoy doing it, and because you love doing it. If you have an impact on society, the money follows.”“Advance your college education as far as you can. I know there are people that say, ‘You don’t even need a college education,’ but statistically, that’s stupid. The average salaries paid to STEM professionals blow away every other field. You don’t look at the one-offs in a field. You have to look at the average statistics, because you’re probably not going to be that one-off, you’re most likely going to end up in the averages. When you’re betting on your career, you want to go into fields with the largest probability of success.”“Stay off social media. Social media can be very damaging. And it’s just a huge time sink. You end up going down rat holes that are totally useless.”“If you are fortunate enough to be successful, don’t forget to give back. One of the most rewarding things you can do in your life is giving back to those less fortunate than you.”

In 1991, very few people had Internet access. Those who did post in online forums or email friends from home typically accessed the Internet via telephone line, their messages traveling at a top speed of 14.4 kilobits per second. Meanwhile, cable TV was rocketing in popularity. By 1991, sixty percent of U.S. households subscribed to a cable service; cable rollouts in the rest of the world were also picking up speed. Hypothetically, using that growing cable network instead of phone lines for Internet access would dramatically boost the speed of communications. And making cable TV itself digital instead of analog would allow cable providers to carry many more channels. The theory of how to do that—using analog-to-digital converters and digital signal processing to translate the analog waveforms that travel on coaxial cable into digital form—was well established. But the cable modems required to implement such a digital broadband network were not on the mass market. Enter Henry Samueli. In 1985, he had established a multidisciplinary research program at the University of California, Los Angeles, to develop chips for digital broadband. Over the next several years, he and his team created a wide variety of proof-of-concept chips demonstrating the key building blocks of high-performance digital modems. And in 1991, Samueli, along with his UCLA grad student Henry Nicholas, founded Broadcom Corp. to commercialize the technology. Today, the innovations in digital signal processing architectures pioneered at UCLA and Broadcom persist in the digital modems that enable both wired and wireless communications in our devices. For these advances, along with contributions to expanding science, technology, engineering, and math (STEM) education, Samueli is the recipient of the 2025 IEEE Medal of Honor. Henry Samueli Current jobsPhilanthropist, Chairman of Broadcom Inc. Date of birth20 September 1954 BirthplaceBuffalo, N.Y. FamilyWife, Susan; three children; three grandchildren EducationBSEE 1975, MSEE 1976, Ph.D. 1980, all in electrical engineering from the University of California, Los Angeles (UCLA) First jobCashier and stock boy in family’s liquor store […]

Editor’s Note: Henry Samueli is the 2025 recipient of the IEEE Medal of Honor. IEEE Spectrum published this profile of Samueli in the September 1999 issue. With the recent explosion in the popularity of cable and digital subscriber-line modems for high-speed Internet access, the odds are that you will soon have one of these broadband communications devices in your home or office—if you don’t already. If you do, the odds are that the chips inside the modem will have been designed by Broadcom Corp., and be based on digital signal-processing (DSP) architectures conceived by Henry Samueli. Eight years ago, Samueli, then a professor at the University of California, Los Angeles (UCLA), who had been pushing the state of the art of digital broadband communications for more than a decade, joined with his Ph.D. student Henry Nicholas to found Broadcom, now in Irvine, Calif. Their first project was to design the world’s first chips for digital interactive television. Today Samueli holds patents for DSP-based receiver architectures for a number of digital communications transceivers, including ones for cable television, satellite television, Ethernet, and high-bit-rate digital subscriber line services. Plus Broadcom now makes more than 95 percent of the chips that go into U.S. digital cable set-top boxes and cable modems. Such modems are viewed as the foundation for the future of data, voice, and video services to the home. Broadcom also has big chunks of the markets for chips for Ethernet transceivers, high-definition television (HDTV) receivers, digital subscriber line modems (the leading alternative to cable modems), and direct broadcast satellite receivers. How a DIY radio kit launched Henry Samueli’s career Samueli’s path toward becoming one of today’s key players in digital communications started 33 years ago, when he was a seventh grader. Required to take a shop class at his West Hollywood, Calif., junior high school, he selected electric shop. During the term, each student was expected to build a crystal radio from a kit, using a single crystal and an antenna wound on a toilet paper tube. Bored with the prospect, Samueli asked his teacher if, […]