SUCCESS

In this episode, we sit down with Shivant Shrestha of the Indiana University Lilly Family School of Philanthropy to explore the new AI-powered methodology behind the latest Million Dollar List.

Transcript:

Jeffrey Snyder, Broadcast Retirement Network

Well, Shivant, it’s great to see you.

Thanks for joining us on the program this morning.

Shivant Shrestha, MA,  Indiana University Lilly Family School of Philanthropy

Thank you for having me.

Jeffrey Snyder, Broadcast Retirement Network

Really appreciate you coming on, and like I told you before in the green room, I’m always excited to talk about charity. I think charitable donations, and I think Americans are some of the most charitable people. We’re gonna talk in detail about the study, but before, and the work that you all did, but before we get into it, let’s talk about the School of Philanthropy and also the meat and potatoes about the research.

How did you perform it, and who sponsored it?

Shivant Shrestha, MA,  Indiana University Lilly Family School of Philanthropy

So, the research project is called the MDL, or the Million Dollar List. It’s compiled by the Indiana University, Lilly Family School of Philanthropy, who are at work, and it documents publicly announced gifts of $1 million or more. So, while it’s not a comprehensive record of all million dollar plus gifts, it is the nation’s most complete dataset on donation level giving of $1 million or more.

This particular report was sponsored by the Gates Foundation, and we conducted this research with the help of a new AI-assisted methodology. So, artificial intelligence helped collect, scrape, and sort the data from publicly available sources online, media sources, articles, various other sources online.

Jeffrey Snyder, Broadcast Retirement Network

Yeah, so before we get into, I’m very curious about the artificial intelligence or super intelligence, depending on who you talk to. We don’t want to go down that road, but I’m interested, how did artificial intelligence help you and the team? Did it speed up the going through the data?

How was it meaningful in terms of the research you provided?

Shivant Shrestha, MA,  Indiana University Lilly Family School of Philanthropy

Sure. So, before this phase of the research, this research covers 2021 to 2023. Back then, the million dollar list data was compiled from IRS Form 990s, media reports, and it was all researched by the school through humans.

So, historically, the data was collected through the researchers at the school, reviewed news articles, press releases, issued by donors and recipients, and other manually entered relevant information into the database. So, while this was still effective, this approach required significant time and human resources and was still not able to capture the full picture due to the limited scope of media coverage or just the time requirements on our researchers. So, starting with this new data, the data collection was modernized through the use of an AI-driven model.

So, it not only sped up the initial data collection process from these sources, the AI was able to broaden and expand the search for these sources, but it also helped track maybe other sources that humans could not have tracked because the internet has grown, there’s just so much information out there.

Jeffrey Snyder, Broadcast Retirement Network

Yeah, I mean, without getting into the details, I think it would actually make the information, the study probably more accurate. So, I applaud you for doing that. And of course, someone is doing the checking, right?

I mean, you don’t just do it and just take it for granted, you have to actually go through and validate and plug it into your own way of actually conducting the research. Let’s talk about some of the key takeaways here. I thought this was really interesting.

First of all, a million dollar donation, that’s pretty generous and it’s very sizable. About how many donations are we talking about here? I mean, the number of, what’s the number that we’re talking about here?

Is it in the tens of thousands, hundreds? What are we talking about?

Shivant Shrestha, MA,  Indiana University Lilly Family School of Philanthropy

From 2021 to 2023, the million dollar list recorded 6,003 donations of a million dollars or above, which amounted to around a total of $109 billion for those three years. So, you’re looking at quite a sizable amount.

Jeffrey Snyder, Broadcast Retirement Network

Yeah, I was gonna say that’s not a small chunk of change. What types of organizations or what types of, these are all individuals, right? But what, these are obviously people that are, I guess, have the resources to do this.

Where are they allocating their monies? Is it in research and technology? Is it in medical research?

Do you have an idea of where people are making their donations? I’m sure they have lots of different interests.

Shivant Shrestha, MA,  Indiana University Lilly Family School of Philanthropy

Sure, so actually the million dollar list captures donations from four different sources. Individuals, foundations, so charitable foundations, whether that’s private foundations, family foundations, corporate foundations, and also corporations. So, major corporations, but also businesses, different forms of them.

Finally, we have another category called other, which is new forms of philanthropy. So, it might be a pooled funds or intermediaries. So, people giving their money to these organizations that are then directing their resources to others as they see fit.

And, but you’re right, individual donors. So, families, individuals in the US constituted the largest source of million dollar plus giving, giving a total of 48.5 billion across over 2,500 gifts during the three years.

Jeffrey Snyder, Broadcast Retirement Network

Okay, so that’s, I guess, the GDP of a small country. That’s pretty powerful, you know, anecdotally, of course. But I think it, what is your takeaway from the study?

I mean, is it that Americans continue to be very charitable, especially maybe more affluent Americans giving back into areas of interest? What are some of the key takeaways here?

Shivant Shrestha, MA,  Indiana University Lilly Family School of Philanthropy

Sure, so I think I’ll break this down in two different points. So, going back to your earlier questions, most of this giving was directed to a sector called human services. So, this would include food banks, emergency services, shelters, different other kinds of services provided in the community and society.

And they receive both the, human services receive both the largest number of gifts, around 1,300 gifts, and also the highest value of them. So, around $23 billion, which was followed by public and societal benefit, and then by higher education. So, that’s one of the big key takeaways that the sub-sectors that we’re seeing, it aligns with other broader research conducted by the school, that these are the sub-sectors receiving kind of the bulk of the charitable giving in the US.

And the second point, which you pointed out again, is that, yes, affluent Americans give, are very generous, but we’re seeing this phenomenon called dollars up, donors down, which means that on average, American households are donating less, but the households that are giving, are giving more.

Jeffrey Snyder, Broadcast Retirement Network

Okay, and so how does that, and I know this, maybe this is kind of, not aligned directly with the topic, but how does this information instruct not-for-profit organizations, or does it? Does it provide, to your last point, does this help them, these not-for-profits kind of frame their ask? Because I would think you want these big donations, but you probably also want smaller donations.

It all adds up, it’s all helpful. So how can, are not-for-profits using, or leveraging this information to streamline how they ask for donations?

Shivant Shrestha, MA,  Indiana University Lilly Family School of Philanthropy

So the Million Dollar List has proven to be a valuable resource for several audiences. So fundraisers and nonprofit professionals are using it for prospect research, to identify donors, to identify affluent Americans who like to support certain causes maybe, or are just broad donors who want their funds to go towards supporting different organizations and different causes. It also helps academics and researchers in the academic space study trends in major philanthropy.

It gives journalists and the general public free access to a record of charitable donations. So it’s being used in many different ways, but certainly nonprofit professionals and nonprofit organizations can use this information to identify donors, understand what kind of trends are ongoing, to see maybe which donors are giving to certain causes, how are they giving, also the timeline. So this report looks at a pivotal time in American history, the world history, actually, 2021 to 2023.

We were facing, we were in the middle of an ongoing COVID-19 global pandemic. There was significant economic and socioeconomic disruption. So it tells us how donors were responding at that particular time, where they wanted their funds to go to support certain causes or certain things, where they wanted to see impact during this time.

Jeffrey Snyder, Broadcast Retirement Network

So let’s talk about some next steps in terms of the research. We’ve talked about the million dollar list, but how do you build, artificial or super intelligence is always iterative, it’s iterative, it’s always learning. So how do you build on what you’ve done?

I think you mentioned that this is the first time that you’ve used large language, not you, but the team has used large language models to sift through the data. Could you see an expansion into different categories? How would you envision this data being used in the future?

Shivant Shrestha, MA,  Indiana University Lilly Family School of Philanthropy

Sure, so definitely, I think the biggest difference we’ve seen through this new methodology is the scale, speed and consistency of the data that we’ve collected. We probably captured gifts that weren’t reported prominently in the media or were buried in large volumes of data. So while it can scan and connect information across sources much more efficiently, at the end of the day, manual collection or human review is definitely required as you pointed out earlier, because at the end of the day, AI is built by humans.

It still makes mistakes, it has a mind of its own. So at the end of the day, human review is still required to kind of sift through the information, make sure what is being collected and displayed to us is accurate and also to identify the broader patterns from it. So we still need to see where this methodology and AI in general will take us, but it doesn’t replace human verification, but it does allow us to spend more time validating and understanding the data rather than trying to find it.

Jeffrey Snyder, Broadcast Retirement Network

Yeah, I would imagine, and I’m not trying to dump more work on you, but you could actually maybe speed up the delivery of a report. Maybe instead of annually, it could become quarterly. Again, I’m not trying to dump on you, but I think having more data, as you said, having more data in real time probably would be very helpful to assessing the trends.

I guess, it’s really interesting, Siobhan, if we take a step back and maybe we’ll conclude here, I wanna get your perspective. Obviously you’re a researcher, but artificial intelligence really has permeated every part of society. I wanna get your reaction to that.

And you’re in an academic setting, so I’m sure you’re seeing it all over the place in the classroom, in the work that you do, et cetera.

Shivant Shrestha, MA,  Indiana University Lilly Family School of Philanthropy

For sure. I mean, it’s become, as you mentioned, it’s permeated every level, every sector that we see today. But I think it’s important to make sure that the human aspect of it remains.

There’s, and also to make it work for us, make it work with us, rather than kind of hand off all of the data collection or analysis to artificial intelligence.

Jeffrey Snyder, Broadcast Retirement Network

Yeah, really well said. Well, Siobhan, we’re gonna have to leave it there. It’s great to see you.

Thanks for making a few minutes. Good luck on the future research, and we look forward to having you back on the program again very soon.

Shivant Shrestha, MA,  Indiana University Lilly Family School of Philanthropy

Sure, thank you so much, Jeff.

Every investing rule carries an assumption about inflation, whether explicitly stated or not. Most of the rules working Americans follow were written when that assumption cost nothing, because inflation sat near 2% and nobody had to price it.

This year, it costs something. Regular gasoline averaged $4.465 a gallon in the week ending Sept. 28, up $1.347 from a year earlier, according to the Energy Information Administration.

Consumer confidence just fell to its weakest reading since 2014. That puts a hole in one of the most repeated rules in personal finance, and the hole is worth more than a percentage point.

Dave Ramsey tells investors to skip bonds

Personal finance expert Dave Ramsey offers guidance on the Ramsey Solutions webpage.

“We don’t recommend betting your retirement on bonds. You’re better off investing your money in a mix of growth stock mutual funds,” reads the page explaining bonds, published in November 2021 and last updated in September 2023.

The page gives two reasons. Bonds “barely outpace inflation,” and if “interest rates are low and inflation increases, inflation could outpace the return.”

Related: Dave Ramsey has surprising advice for homebuyers

The guidance also carries two links to SmartVestor Pro, the advisor-referral network Ramsey Solutions discloses it promotes for a fee, which is worth knowing when a rule steers you out of an asset class.

His broader philosophy, last updated on his site in April 2026, puts 15% of income into four types of growth stock mutual funds and assumes stocks return 10% to 12% a year.

On the median household income of $87,460, that 15% works out to $13,119 a year, or $1,093.25 a month. TheStreet has examined the 8% withdrawal rate he pairs with those returns.

Dave Ramsey tells investors to skip bonds for growth stock mutual funds.Anna Webber / Getty Images

Why one Treasury breaks Ramsey’s bonds rule

The Treasury sells two kinds of notes. One pays a fixed number, and the other pays a fixed number on top of whatever inflation turns out to be, because its principal moves with the consumer price index.

Ramsey’s page does not separate them, and the difference is the whole story. On Sept. 28, the 10-year Treasury note yielded 5.24%, while the 10-year inflation-linked note yielded 2.90% above inflation, according to the Treasury Department‘s own daily yield curves.

More Dave Ramsey:

Dave Ramsey emphasizes key warning on 401(k)s, IRAs

Dave Ramsey says this insurance product is horrendous for families

Dave Ramsey just flagged 3 retirement mistakes Americans make

Subtract one from the other, and you get the inflation rate the bond market is pricing over the next decade, which is 2.34%.

That 2.90% real yield stood at 2.44% on Aug. 31, so it climbed 46 basis points in four weeks.

A $19 billion reopening on Sept. 17 cleared at a 2.653% real yield, the highest for that term since October 2008, TipsWatch publisher David Enna reported that day. TheStreet covered real yields climbing past 2% in July, and they have gone higher since.

Households and the market disagree sharply

Consumers put 12-month inflation at 6.1% on average and 5.1% at the median in September, both up three tenths of a point, the Conference Board reported Sept. 29.

Its Consumer Confidence Index fell 6.7 points to 81.9, against a consensus near 89.2. The Expectations Index dropped to 63.6, a third straight monthly decline and well under the 80 level that has historically flagged recession risk.

“References to prices, the high cost of goods and services, and oil and gas prices in particular, rose to new heights, reflecting September’s surge in fuel costs,” said Dana Peterson, chief economist at The Conference Board.

Take the household number at face value, and the arithmetic turns hostile quickly. A 10-year Treasury at 5.24% loses 0.81% a year in purchasing power at 6.1% inflation, so $10,000 comes back worth $9,218 in today’s money.

The inflation-linked note earns 2.90% above inflation whatever inflation does, turning that same $10,000 into $13,309 in today’s money. The spread is $4,091 on a single $10,000 decision.

At a glance

10-year Treasury note, Sept. 28: 5.24%

10-year inflation-linked note, Sept. 28: 2.90% above inflation

Implied 10-year inflation rate: 2.34%

Consumer 12-month inflation expectation: 6.1% average, 5.1% median

Consumer Confidence Index, September: 81.9, down from 88.6

Expectations Index, September: 63.6

Regular gasoline, week ending Sept. 28: $4.465 a gallon

Top national savings rate: 4.20% APY against a 0.65% averageSources: Treasury Department daily yield curves; The Conference Board; Energy Information Administration; Bankrate. The implied inflation rate is TheStreet’s calculation.

The number that squeezes Ramsey’s case

Run the comparison the way an investor actually faces it. A 2.90% real yield is worth the same as a 9.18% nominal return if inflation runs at the 6.1% households expect.

Ramsey’s own page claims stocks return 10% to 12%. At the top of the household expectation, his advantage over a government note carrying no market risk narrows to 0.82 percentage points.

At the Fed’s own projection of 3.4% core inflation for 2026, that same 2.90% real yield is worth only 6.40% nominal, and his advantage widens back to between 3.60 and 5.60 points.

The distance between those two outcomes is the entire argument, and it turns on an inflation number nobody has yet.

Why households may be wrong anyway

Consumers have badly overshot before. Their inflation expectations were more unanchored in early 2025 than at the late 1970s peak, once you account for how widely their forecasts disagreed.

That finding comes from Robert Rich and Alexander Cline of the Cleveland Fed‘s Center for Inflation Research, in work published Feb. 2. One-year-ahead expectations peaked above 9%.

The same bank reported in March that anchoring “deteriorated notably for consumers in 2025,” while professional forecasters showed no such weakening. So 6.1% reads as a fear number more than a forecast.

Inflation is still doing real damage at far lower rates. Real home values fell for a 14th consecutive month in July, even as nominal prices rose 1.9%, according to the S&P Cotality Case-Shiller index released Sept. 29.

What it means for you

Inflation-linked notes sell in $100 multiples in 5-, 10-, and 30-year terms, and at maturity, you collect the greater of your original or inflation-adjusted principal, according to NerdWallet.

Hold them inside a 401(k) or IRA where you can. The inflation adjustment counts as federal taxable income in the year it accrues, even though no cash reaches you. It does, however, escape state and local tax.

Ramsey’s 15% savings rate holds up, and so does his warning about concentrating in single stocks. The line to check is the one treating every government bond as the same instrument.

If you already follow his allocation, the useful question is whether the fixed-income slice you were told to skip now pays enough to fund a few years of retirement spending without market risk attached. His stance on holding stocks deep into retirement makes that question sharper, not softer.

The next reading of household inflation expectations lands Oct. 27, and the Fed meets before that. Both will move the number this decision turns on.

Related: Dave Ramsey warns Americans on 401(k)s, Roth 401(k)s

TheStreet aims to feature only the best products and services. If you buy something via one of our links, we may earn a commission.

Why we love this deal

Fall is prime movie-watching time. With cooler weather and the cozy season in full swing, one of the best ways to end the day is with a movie night, whether it’s solo or with friends and family. For me, I’m prioritizing all things Halloween, with everything from animated classics to movies that just landed on streaming services. But the one thing that would make any movie night better? A projector that gives you the feeling of a home movie theater. 

Ahead of Amazon’s Prime Big Deal Days is the perfect time to snag one. The Elephas Roku TV Smart Projector is currently on sale for only $100, which is 23% off its regular price of $120. With projectors often costing $200 or more, it’s a fantastic deal. However, it’s only available for a limited time, with only a few more hours left on the clock.

Elephas Roku TV Smart Projector, $100 (was $130) at Amazon

Courtesy of Amazon

Shop at Amazon

Why do shoppers love it?

If you’ve ever wanted a movie theater experience at home, this projector can make it happen. With built-in Roku TV, all you have to do is sign in, and you can watch TV shows and movies from your go-to streaming services with ease. As if it wasn’t easy enough to set up and use, it also comes with a remote control to make navigating it even easier.

The projector is a smart device with all the bells and whistles you could possibly need. It’s compatible with Siri, Alexa, and Google Assistant, making it responsive to voice commands, whether you need to adjust the volume or pause or play. You can also use Apple AirPlay to display and share photos, videos, and music.

It also has a built-in speaker and a clear picture. You get Dolby Audio, as well as manual focus for screens ranging from 40 inches to 150 inches. The projector also has WiFi for a stable connection and Bluetooth connectivity to connect to additional speakers or headphones. 

At only 6.81 inches long by 5.55 inches wide by 2.48 inches high, it’s super compact, too, which makes it portable and easy to move around. You can take it from room to room, or you can set up your very own outdoor theater on your patio or backyard.

Related: Amazon has a ‘clear and responsive’ smartwatch for just $20 ahead of October Prime Day

Details to know

Dimensions: 6.81 inches long by 5.55 inches wide by 2.48 inches high.

Weight: 1.28 pounds.

Features: Roku TV, Dolby Audio, Wi-Fi, Bluetooth, built-in speaker.

Amazon shoppers praise this projector on practically all fronts. Reviewers rave about everything from its picture quality to its overall convenience. “Having Roku built in makes it very convenient because I can access my favorite streaming apps without needing another device,” a customer said. “The picture quality is so good for a compact projector, and it’s easy to move around and set up.” Shoppers also note that the built-in speaker is a win, as it provides “clear audio.”

Shop more deals

Zdk Smart Projector, $98 (was $200) at Amazon

Goodee 4K Smart Projector, $240 (was $400) at Amazon

Aurzen Roku TV E1R Smart Mini Projector, $90 (was $140) at Amazon

The Elephas Roku TV Smart Projector is on sale for $100, but with a limited-time deal, it won’t be for much longer.

The debate over what artificial intelligence will do to workers has been running for years.

Companies argue that the technology will free employees from repetitive tasks and open space for more meaningful work. Critics warn that automation will eliminate jobs faster than new ones can appear. Both sides have data, and neither has settled the argument.

What has shifted is who is weighing in, and where. The executives building AI systems, the investors funding them, and the government officials overseeing them are increasingly in the same room.

The conversations are becoming more specific because the technology is no longer theoretical. It is already being deployed in customer service, software development, financial analysis, and content production at scale.

That backdrop framed a recent meeting at the White House, where some of the most powerful figures in American business and technology gathered to discuss AI and its economic consequences.

What one of them said afterward was brief. It was also difficult to ignore.

Also read: Elon Musk joins Tim Cook in sending strong warning to Americans

What Elon Musk said about jobs and AI

After a White House meeting with President Donald Trump and other artificial intelligence executives, SpaceX CEO Elon Musk delivered a powerful six-word message, as CNBC reported: “AI will significantly alter the workplace.”

He stopped short of predicting mass unemployment but left little room for the idea that most jobs would remain unchanged. Workers are likely to see their roles evolve as AI systems grow more capable, he said.

That framing carries weight because of who is saying it. Musk leads companies that are both deploying AI and building the infrastructure behind it. His businesses use AI for manufacturing optimization, autonomous driving, customer communication, and software development. He is not speaking as an outside observer.

The meeting covered AI’s broader economic impact and raised questions about the responsibilities of companies developing the technology, ABC News reported.

The discussion came as AI policy is beginning to produce real regulatory consequences in Washington and in state capitals across the country. What companies are required to disclose, test, and guard against before releasing AI products is an active and unsettled question.

How AI is already changing what workers do every day

AI systems can already handle writing, coding, research, customer service, data analysis, and image generation at a level that competes with entry-level human output. That is not a forecast. It is the current state of tools that millions of workers and businesses are already using.

The immediate effect on employment is uneven. Some roles are disappearing. Others are being redesigned so that workers manage AI systems, review their output, or concentrate on tasks requiring judgment and human interaction.

A customer service team may shrink while the remaining agents handle more complex cases. A software team may produce more work with fewer developers.

Workers whose jobs consist mainly of predictable, repeatable tasks face the most immediate pressure. Roles built around processing information, drafting standard documents, or completing structured data work are the most exposed to automation.

Those are also among the most common job categories in the American economy.

Musk acknowledged that AI will significantly alter the workplace.Luis Alvarez / Getty Images

How companies plan to use AI to cut costs

For businesses, the financial case for AI adoption is direct.

A company that uses AI to handle routine customer inquiries may need fewer support agents. A software firm may allow developers to complete more work with automated coding tools. Professional services firms are already using AI to process documents, prepare initial analysis, and summarize large volumes of information.

Those changes can improve profit margins. They can also reduce demand for certain types of labor. The net economic effect depends on whether new AI-related tasks and industries generate enough employment to offset what automation removes.

That equation has not resolved itself. Productivity gains from past waves of automation eventually produced new categories of work, but the adjustment took time and was unevenly distributed. AI could follow the same pattern, or it could move faster than previous technological shifts.

Who is liable when an AI system makes a costly mistake?

The White House meeting also focused on accountability. AI executives signed a voluntary commitment to self-police their technology, CNN reported.

AI systems can make decisions, generate content, and carry out multistep tasks with limited human supervision. When those systems produce errors, the consequences can include financial losses, privacy violations, and discrimination.

More Elon Musk:

Elon Musk makes bizarre claims about money, future of AI

Elon Musk sends blunt verdict on the future of humanity and AI

Elon Musk’s startling claim to SpaceX investors

Technology companies may face growing expectations to test their products thoroughly, disclose limitations, and build safeguards before deploying AI to consumers and businesses.

The legal frameworks for assigning responsibility when AI causes damage are still being written. Regulators in the United States and Europe are working on rules that would require developers to document how their systems operate and the risks they carry.

Companies that deploy AI in high-stakes areas, including hiring, lending, and medical diagnosis, are under the most scrutiny. Errors in those contexts carry consequences that go beyond customer complaints.

The skills that hold their value as AI takes over routine work

Workers who develop abilities that complement AI rather than compete directly with it are better positioned for what follows.

Critical thinking and professional judgment are harder to automate than information retrieval or document drafting. Communication, relationship management, and industry-specific expertise hold their value when AI handles the routine layer of a job.

Understanding how to evaluate AI-generated output is becoming a practical requirement in many fields. Knowing when a result is accurate, when it is plausible but wrong, and when it needs human review is a skill that does not come built into the tools. Workers who develop that judgment alongside technical fluency are likely to remain in demand.

Companies face their own adaptation challenge. Integrating AI into existing workflows, protecting sensitive information, and building procedures to check AI output all require investment and planning. Businesses that move fast without those safeguards take on operational and reputational risk that is easy to underestimate.

What higher AI productivity actually means for paychecks

Productivity gains from AI could give workers more time for strategy, innovation, and client relationships. They could also allow smaller businesses to access capabilities that previously required specialized staff at larger organizations.

The distribution of those gains is not guaranteed. Productivity increases do not automatically translate into higher wages or better conditions. Companies have discretion over how efficiency gains are allocated, whether to shareholders, customers, employees, or reinvestment.

That is a question of business decisions and policy choices, not of what the technology can do.

Musk’s remarks at the White House reflect something that most people in the industry now accept. The job market is changing because of AI.

How fast that change moves, and who benefits from it, remain open questions that neither executives, regulators, nor workers have fully answered yet.

Related: Goldman Sachs sends strong message on AI and jobs