Suggestions
Sal Parsa
Project Leader @ BCG - Berkeley MBA - Ex-founder/CEO @ Predictim
Sal Parsa is an accomplished professional with a strong educational background and diverse work experience.
Sal holds a Master of Business Administration from the prestigious Haas School of Business at the University of California, Berkeley. He also completed the SkyDeck Accelerator program and earned a B.A. in Business Management and International Relations from Walsh University.
His career includes roles at esteemed organizations such as Boston Consulting Group (BCG) where he served as a Project Leader and Consultant. Additionally, Sal held positions at EY-Parthenon as a Consultant, founded Predictim, and interned for Congresswoman Betty Sutton at the U.S. House of Representatives.
Sal's professional journey also led him to work in communications and public relations at The Timken Company and as an Analyst/Interpreter for the US Government.
His diverse experience across consulting, entrepreneurship, politics, and corporate environments showcases his adaptability and skill set.
Highlights
We had Sam Altman on campus, as part of an event organized by Xfund. His remarks on ChatGPT/AI/AGI were extremely insightful. My takeaways:
- I asked him if OpenAI might consider other strategies to monetize LLMs, beyond subscriptions - potentially spurring innovation powered by GPT-4/5, similar to how Google monetized search with ads, helping to create other services like Gmail and Maps. His response was that ads were important to give the early internet a business model, but ads and AI is uniquely unsettling. He hats ads but they haven't been fully rolled out.
- Setting the appropriate boundaries for ChatGPT’s behavior, what it should and shouldn’t do, remains a huge challenge.
- AI is like a "scaffolding of shared intelligence” that can merge and reassemble knowledge - it will accelerate innovation and broaden our collective wisdom.
- AGI will have overwhelmingly positive impact.
- Cost of cognitive labor could drop significantly, by a factor of a million or even a billion; the limited resources then will be compute and energy, both capital-intensive.
- People often underestimate the rapid advancements of AI models because of linear thinking habits (vs. exponential); they overlook the capabilities and the potential impacts of future models like GPT-5 and GPT-6.
- Advice for entrepreneurs: Don’t think of the current AI capabilities as the limit; build products with an eye toward the future and the capabilities that newer models will bring - imagine having a team of brilliant engineers constantly upgrading your tech stack. #Harvard #AI # #Innovation #Entrepreneurship
Over $150 billion has been invested in Afghanistan's reconstruction efforts—surpassing the inflation-adjusted amount spent on the Marshall Plan in Europe after World War II. Despite this, Afghanistan's GDP per capita remains nearly the same as it was two decades ago. How is this possible?
I am excited to moderate this upcoming event for the Harvard Afghanistan Caucus, featuring Inspector General John Sopko (SIGAR). He will share insights into the successes and challenges of America's largest reconstruction project and lessons learned.
It’s a privilege to be able to bring such pivotal discussion to the community at Harvard Kennedy School.
#harvard #afghanistan
