The case for science and technology research and development funding
Summary according to ChatGPT
- OpenAI’s ChatGPT and similar AI products by Microsoft and Google have raised concerns and excitement about their impact on society. There is a need for a partnership between the government and the private sector to ensure responsible development and use of these tools.
- Increased federal funding is necessary to boost innovation, counter the trend of a stagnant economy, and address growing income inequality. The authors of Jump Starting America recommend targeted investments in specific regions, such as the greater Syracuse area, which has many favorable factors for leveraging such investments.
- The Chips and Science Act, passed in 2022, has resulted in federal funding targeting the development of the semiconductor industry, which aligns with the vision outlined by Johnson and Gruber.
The release of OpenAI’s ChatGPT has brought Artificial Intelligence to the forefront of the public’s attention. With Microsoft and Google also developing their own AI products based on OpenAI research, there is both excitement and concern about their potential impact on society. While these tools offer opportunities for economic growth, it is important that a partnership is established between the federal government, local and regional governments, and the private sector to ensure their responsible use and development.
Last week, Kara Swisher had Sam Altman, the CEO of OpenAI, as a guest on her podcast “On.” During the podcast, Swisher asked Altman several questions about OpenAI’s technology. I highly recommend listening to the entire podcast. One of the questions Swisher asked Altman was why OpenAI switched from a nonprofit to a for-profit model and has since raised $11 billion, according to Crunchbase. Altman explained that developing this technology is expensive, and the nonprofit model did not provide the economic incentives needed to fund the organization. To finance the technology, they needed cash from investors. Altman referred to their model as a “cap-profit” model, where profits beyond a certain amount will go to the non-profit arm for research and development.
Swisher inquired about who else might have funded OpenAI, to which Altman replied that historically, the US federal government would have supported and developed such technology. Despite OpenAI’s attempts to secure federal funding, the funds were unavailable. In contrast, China and other countries receive federal funding or support for developing large language models and other forms of artificial intelligence. While the private sector drives innovation in the US, relying solely on this approach for research and development poses a risk and does not align with the science and technology R&D of the mid-20th century.
In her other podcast, Pivot, Swisher discusses her reflections on the Altman interview in more detail, and it’s definitely worth a listen. Co-host Scott Galloway hopes that the individuals leading the further development of this technology are ethical and moral, given that their influence will be akin to that of Bill Gates, Elon Musk, and possibly Steve Jobs. These individuals’ impact extends not only to technology but also to their approach to work and life.
Microsoft’s Bing recently integrated with GPT-4, prompting New York Times reporter Kevin Roose to share a concerning experience he had with the bot when he tested it. OpenAI and Bing responded by placing limitations on the technology to avoid creating interactions that appear to be sentient. Altman, in an interview, argues that releasing products to the public is crucial for identifying and addressing issues, ultimately improving the product.
Facebook released its large language model LLaMa with the intention to first only authorize specific types of users — academics and researchers — hoping to avoid some of the negative attention that came to ChatGPT and Bing’s Sydney. Almost immediately, though, someone released the code online and it is now available to find. In the meantime, students at Stanford trained their own large language model for $600 and open sourced it. Galloway’s hope that the people developing these tools are good gets more challenging when more and more people have access to the technology and can built their own versions. There were initial claims that ChatGPT had a left-leaning political bias, and in recent weeks, efforts to build right-leaning specific versions of ChatGPT have begun.
Technology has rapidly become ubiquitous, and it is impossible to undo its widespread adoption. The current discussion has shifted towards regulating this technology. Altman has expressed his belief that regulation is necessary, although he is uncertain about the extent of its implementation. If the US government had sponsored research and development into this form of artificial intelligence, how might the situation be distinct in an alternate reality?
According to Jump Starting America, a book by Jonathan Gruber and Simon Johnson, the United States’ federal investment in science and technology research and development from World War II to the late 1980s helped the country stay ahead in innovation and build a thriving middle class. Specifically, this investment fueled the space race, which helped establish the US as a leader in space exploration and technology.
In the decades that followed World War II, the U.S. led the world in innovation, creating entirely new sectors such as jet aircraft, life‐saving drugs and vaccines, microelectronics, satellites, and digital computers. Widespread innovation boosted productivity. Household income increased faster than ever before, while inequality declined.
Over the last two decades, federal funding for R&D has seen a sharp decline, leading to a stagnant economy and growing income inequality. Moreover, large-scale innovation has been slow to materialize, and wealth has become concentrated in specific regions of the country. To counter this trend, the authors of a book on science and technology argue that increased federal funding is necessary, with targeted investments in specific regions holding great potential.
The authors have identified several areas of the country that are particularly well-suited to leverage such investments, with the greater Syracuse area being among the top locations. This is due to factors such as the population’s education level, proximity to research institutions, low cost of living, and convenient transportation options. As someone who worked for the city, reading this book was particularly exciting. In fact, our own Syracuse Surge strategy aligned with many of the book’s recommendations.
In recent years, the Chips and Science Act, passed in 2022, has resulted in federal funding that specifically targets the development of the semiconductor industry. This funding aligns with the vision outlined by Johnson and Gruber. Thanks to the combined efforts of local, state, and federal officials, as well as community members and organizations, Micron has committed to investing $100 billion to build a semiconductor plant in the Syracuse area. This is one of the largest investments in the history of the United States and it will be located right here in Central New York.
This investment in semiconductors is strategic in nature because the vast majority of semiconductors are currently produced in Taiwan. The need for this technology has grown so great that any disruption in the supply chain can have a significant negative impact on the economy. By bringing semiconductor production closer to home, we are ensuring the future stability of our economy.
In their paper following up on Jump Starting America, Johnson and Gruber argue that having more scientists in a city can lead to a greater increase in productivity than the increase in costs. Investing in research and development can have a positive effect on a region, and the recent investment by Micron in Syracuse is hoped to jump-start the economy of Central New York, which has been struggling for decades. However, there are challenges that need consistent investment. According to a Wall Street Journal article, training enough workers in the next decade to meet the demand created by Micron will be a daunting task that will require universities and colleges to shift their focus. To address this need, a Science, Technology, Engineering, Arts, and Math (STEAM) high school, which was proposed when I worked for the City, is now being established to prepare high school students for future employment at Micron. Additionally, the Museum of Science and Technology, on whose board I serve, has received funding from Micron to promote education in STEM fields for kids. We need to ensure that our students are prepared for such work. It is exciting to see dedicated efforts being made towards this goal, but the risks are also significant. Such investments should have been made earlier.
The Syracuse Surge initiative aims to address the issue of poverty in the City of Syracuse by focusing on education, job readiness, housing stability, and internet connectivity. Recently, the passage of the Chips bill and Micron’s investment have provided a boost to this plan, accelerating the need for training and development in the area.
I recently had a conversation with a colleague about OpenAI and its potential impact on the future. While we were excited about the possibilities, we also expressed concern that the technology could be used to eliminate jobs and increase efficiency at the expense of people’s well-being. The Syracuse Surge strategy, along with the arguments of Johnson and Gruber, suggests that technological advancements can benefit everyone, not just a select few. However, as Susan Crawford notes in her book Fiber, unequal access to critical infrastructure such as fiber internet can leave some behind. Furthermore, some of the most successful companies today prioritize saving time for the wealthy by paying low wages to “contractors” for driving or food delivery, perpetuating the problem of inequality.
With the rise of technologies like OpenAI, there is a growing concern that robots could replace human jobs. Simon Johnson and his collaborator argue that professions like online and phone customer service agents may soon be entirely automated. However, rather than focusing on the risk of job loss, we should prioritize enhancing the abilities of nurses, teachers, and customer service agents to better understand and serve their clients and customers. By leveraging the collective knowledge of human history, we can push ourselves to provide better work and services, benefiting all of humanity.
This shift in focus requires support from the federal government, which must consider how to facilitate the continued development and regulation of these technologies. The goal should not be to stifle innovation but to optimize it and ensure that everyone benefits from it. Without government support, we rely on private investors to recognize the potential of these technologies and support them, and we hope that company CEOs will ensure their safe and secure growth.
As an AI, data, and tech enthusiast, I believe that this technology marks an inflection point where everything is set to change. However, to ensure that this change is positive, we need federal investment, training in school districts supported by states and local governments, and support for local economic development that leverages and implements these technologies to create opportunities for all and grow the economy. This support should be ongoing, rather than waiting for the next transformative technology to emerge.