April 12, 2024

Women in AI: Kathi Vidal of the USPTO has been working on AI since the early 1990s

To give AI-focused female academics and others their well-deserved (and long overdue) time in the spotlight, TechCrunch is launching a series of interviews focused on notable women who have contributed to the AI ​​revolution. As the AI ​​boom continues, we will publish several pieces throughout the year, highlighting important work that often goes unrecognized. Read more profiles here.

Kathi Vidal is an American intellectual property attorney and former engineer who serves as director of the United States Patent and Trademark Office (USPTO).

Vidal started her career as an engineer for General Electric and Lockheed Martin, where she worked in AI, software engineering and circuits. She holds a bachelor’s degree in electrical engineering from Binghamton University, a master’s degree in electrical engineering from Syracuse University, and a JD from the University of Pennsylvania Law School.

Question and answer

In short, how did you get started with AI? What attracted you to the field?

When I started studying at the age of sixteen, I was interested in solving scientific problems. I had an oscilloscope that I bought at a garage sale that I tinkered with all the time, and I loved working on my Dodge Dart! This early fascination led me to GE’s Edison Engineering Program as one of two women selected for the program. On a weekly basis, we were involved in solving technical problems in all technical and scientific disciplines, in addition to alternating work assignments in different technical areas. When I was approached to work in a three-person team in the field of artificial intelligence, I took the plunge. The ability to do new, groundbreaking work in the early 1990s that could be applied across all scientific and engineering disciplines to devise ways to innovate more creatively was exciting. I saw it as a way to break away from the rigidity of current design principles and to better emulate the nuances that people bring to solving problems.

What work are you most proud of (in AI)?

It would connect my current work on US government AI policy at the intersection of AI and innovation, and my work developing the first AI fault diagnosis system for aircraft. Regarding the latter, in the early 1990s I worked with neural networks, fuzzy logic and expert systems to build a resilient, self-learning system. Even though I went to law school before the system was deployed, I was excited to create something new in the relatively nascent AI space (compared to where AI is now) and to work with the PhDs at GE Research to advance the lessons in share our projects. . I was so excited about AI that I ended up writing my master’s thesis about my work.

How do you address the challenges of the male-dominated technology industry, and, by extension, the male-dominated AI industry?

To be honest, the way I dealt with technical challenges in the 1990s was by conforming (without realizing I was conforming). It was a different time, and it probably goes without saying that most leadership positions in tech and law firms were more male-dominated than they are today. Some of my male colleagues suggested to me that I should learn to laugh less. But I found joy in life and what I did! I remember speaking to a room full of women at a women’s conference we founded in the mid-2000s (before women’s conferences became the norm). When I finished speaking, a number of audience members came up to me to congratulate me on my speech and tell me that they had never seen me so lively and animated. And I was talking about patent law. At that moment I had an ‘aha’ moment; Being appreciated for being authentic was how I felt engaged and successful in my work.

Since that time, I have made a conscious decision to be authentic and create inclusive environments where women can thrive. For example, I have revamped hiring and promotion practices in the organizations where I have worked. Most recently at USPTO, our agency saw a nearly 5% increase in diversity among our leadership ranks within a year as a result of these changes. I have been committed to policies that open the doors for more women to participate in innovation, recognizing that while more than 40% of those who use our free legal services to file patent applications identify as women, only 13% of patented inventors are women. – so we are working hard to close that gap. Together with U.S. Secretary of Commerce Gina Raimondo, I founded the Women’s Entrepreneurship initiative within the U.S. Department of Commerce to empower more female business leaders and arm them with the information and help they need to succeed. as well as other communities historically underrepresented in our innovation ecosystem through my work leading the Council for Inclusive Innovation and the Economic Development Administration’s National Advisory Council on Innovation and Entrepreneurship. In my free time, I also spend time mentoring others, sharing lessons learned, and developing the next generation of leaders and advocates. Of course, I cannot do this work alone – it all happens through and with like-minded women and men.

What advice would you give to women looking to enter the AI ​​field?

First, we need you, so keep going. It is important to involve women in shaping the AI ​​models of the future to limit bias or safety risks. And there are so many pioneers – Fei-Fei Li of Stanford and Elham Tabassi of the National Institute of Standards and Technology (NIST), to name a few. I’m honored to work with incredible leaders at the forefront of AI – Department of Commerce Secretary Raimondo and Zoë Baird, NIST Director Laurie Locascio, Copyright Office Director Shira Perlmutter, and new AI Safety Institute Leader Elizabeth Kelly . It is imperative that we all work together, across government and the private sector, to create the future, or it will be created for us. And it may not be the future we believe in or want.

Second, look for the wind at your back and persevere. Pop the question and put your goals at the forefront to attract others to support you on your journey. Don’t take “no” personally. See ‘no’ and resistance as headwinds. Find your tailwind and the mentors and sponsors who care about you, your success and what you can contribute in this terribly important area.

What are some of the most pressing issues facing AI as it continues to evolve?

The US is fortunate to lead the world in innovation by AI developers, so we also have a responsibility to lead policies that make AI safe and trustworthy and advance our values. We pursue this in collaboration with other countries, in various multilateral locations and bilaterally. USPTO has a long history of this type of collaboration and leadership. To ensure that American values ​​are embedded in AI policy, our AI and Emerging Technology Partnership that we launched in 2022 supports the Biden Administration’s government approach to AI, including the National AI Initiative, to advance America’s promote leadership in AI. We recently published guidance clarifying the level of human contribution required to patent AI-based inventions, advancing human ingenuity and encouraging investment in AI-based innovations, without hindering future innovation by unnecessarily locking innovation in or suppress competition. To our knowledge, this is the first such directive in the world. We need to achieve the same goals and balance when it comes to our creative industries, and we’re working with stakeholders and the Copyright Office to do that.

While we at USPTO focus on leveraging AI to democratize and scale innovation, and on policy at the intersection of AI and intellectual property, we are also working with NIST and the National Telecommunications and Information Administration on other pressing issues, including the safe, secure and reliable development and use of AI and mechanisms that can create deserved trust in AI.

What issues should AI users be aware of?

As President Biden stated in his executive order on AI, responsible AI use has the potential to solve pressing challenges and make our world more prosperous, productive, innovative, and secure, while irresponsible use can worsen societal harms “such as fraud, discrimination, prejudices. and disinformation; displace and disempower employees; suppress competition; and pose a risk to national security.” AI users must be thoughtful and deliberate when using AI so that they do not allow these harms to continue. One important way is to stay informed about the work NIST is doing through its AI Risk Management Framework and its US AI Safety Institute.

What’s the best way to build AI responsibly?

Together. To build AI responsibly, we need not only government intervention and policy, but also industry leadership. President Biden recognized this when he convened private AI companies and gave them voluntary commitments to manage the risks of AI. We at the US Government also need your feedback as we carry out our work. We regularly solicit your input through public engagements and through requests for information or comment that we post in the Federal Register. For example, through our AI and Emerging Technology Partnership, we solicited your comments before we designed our Inventorship Guidance for AI-Assisted Inventions. We are using your comments in response to the Copyright Office’s request for information related to the intersection of copyright and AI to advise the Biden administration on domestic and international strategies. NIST has asked for your input and information to support the safe, secure, and reliable development and use of AI, and NTIA has asked for your feedback on AI liability. And we at USPTO will soon issue another request for comment to explore how our patent laws may need to evolve to take into account how AI could impact other patentability factors or create a minefield of “prior art,” making it more difficult to patent. . The best thing you can do is stay informed about the government’s work on AI, including that of NIST, USPTO, NTIA, and the Department of Commerce in general, and provide your feedback so that together we can create responsible AI can build.

How can investors better push for responsible AI?

Investors should do what they do best: invest in the work. Progress in responsible AI cannot come out of the blue; we need companies in this space to do the hard work to create the responsible AI companies of tomorrow. We need investors to ask the right questions, push for responsible development, and use their money to support the responsible AI of the future. Furthermore, they should convince companies they invest in of the need to prioritize IP protection, cybersecurity and not accepting investments from suspect sources. All three are necessary to guarantee control over work and ensure that work creates jobs and strengthens national security.

Leave a Reply

Your email address will not be published. Required fields are marked *