WASHINGTON — In November 2016, Dipayan Ghosh was still reeling from Hillary Clinton’s defeat as he left what was supposed to be a celebration party at the Javits Convention Center in New York to attend morning meetings for his job at the Washington offices of Facebook.
As Mr. Ghosh, a former White House technology adviser to President Barack Obama, made the four-hour drive, troubling questions started nagging him. What if fake news on Facebook and other sites had an impact on voters? How did the campaigns and any outsiders use ads on the site to influence the election?
A few months later, Mr. Ghosh quit his job at Facebook, where he worked on privacy and public policy issues. On Tuesday, a Washington think tank, New America, and Harvard’s Shorenstein Center on Media, Politics and Public Policy published a report he co-wrote, asserting that technology behind digital advertising — the financial lifeblood of Facebook, Google and Twitter — has made disinformation campaigns more effective.
“The problems were much broader than we imagined, and it was not just about one tool or platform,” said Mr. Ghosh, who with his co-author, Ben Scott, worked on devising Mrs. Clinton’s tech policy platform. “It’s the profit model underlying the whole digital advertising system.”
Mr. Ghosh and Mr. Scott are the latest members of the political party that more eagerly embraced Silicon Valley to sharply criticize the tech industry. Tech policy officials from the Obama administration and from Mrs. Clinton’s campaign, as well as prominent Democrats in Congress, are demanding changes from companies they had long viewed as too important and nimble for regulations.
Senators Mark Warner of Virginia and Amy Klobuchar of Minnesota are demanding greater disclosures from companies like Facebook, Google and Twitter for political advertising on their sites. Senator Cory Booker of New Jersey, who has big political backers from Silicon Valley, fears the biggest companies have edged out competition.
Senator Richard Blumenthal of Connecticut has called for the Federal Trade Commission to restart an antitrust investigation into Google. And Representative Keith Ellison of Minnesota has introduced a bill that would update antitrust policies to take more direct aim at the tech sector.
“Democrats and progressives still strongly feel that there are shared values with Silicon Valley, but there is also a real concern over the industry’s increasingly concentrated wealth and power,” said Daniel Sepulveda, an ambassador and deputy assistant secretary at the State Department for the Obama administration.
Karen Kornbluh, who served as Mr. Obama’s ambassador to the Organization for Economic Cooperation and Development, said that before the election, there was a view of “internet utopianism” in government. Officials looked to the internet to solve problems in education, income inequality and global democracy. They called on tech executives to help with those initiatives and hired from Google and other tech outfits to bring their expertise into the White House.
“So if you were championing the best things about the internet, it was easy to be disappointed that it was hijacked to subvert the very things it could foster,” Ms. Kornbluh said. But she said few people in government were looking with a full view of how social media and other internet services posed national security, economic and other risks.
Mr. Ghosh and Mr. Scott played a leading role in helping to create the tech-friendly policies that helped companies like Facebook and Google flourish during the Obama administration. But as more information trickled out about the role played by technology firms in Russia’s attempts to influence the presidential election, they, like many Democrats, became disillusioned.
“We were always careful to condition our optimism and our advocacy that this technology was potentially a double-edged sword,” said Mr. Scott, who is a senior adviser to the Open Technology Institute at New America. “But I guess we didn’t expect them to hit home quite as hard as it did.”
Titled “#DigitalDeceit: Exposing the Internet Technologies of Precision Propaganda,” their report argues that the interests of internet giants in helping advertisers run persuasive campaigns are aligned with those of someone looking to spread misinformation.
The authors suggest a few ways to regulate the advertising technology industry, including requiring more transparency for political advertising, restricting data collection or ad targeting on political issues, and strengthening consumer protection and competition policies.
Republicans have also expressed concerns about internet giants. Many on President Trump’s campaign suspected that Google rigged search results in favor of Mrs. Clinton, questioning the influence of high-profile supporters of hers like Eric Schmidt, who is the former executive chairman of Google’s parent company, Alphabet. They were outraged by claims that Facebook manipulated its trending topics feature to push down results related to Republicans.
But increasingly, the most vocal criticism is coming from Democrats. Tom Wheeler, Mr. Obama’s head of the Federal Communications Commission, recently called for social media firms to make it easier to figure out how certain information goes viral, who is sending those posts and what impact that activity may have.
“The internet was once heralded as the great democratizing tool,” Mr. Wheeler said. “That vision was smashed by the algorithms of the social media platforms.”
Last year, Google, Facebook and Twitter admitted that groups with ties to the Kremlin had used their online services to push false news stories or controversial headlines on divisive topics. In some cases, those groups amplified their reach by buying ads.
On Friday, Twitter said it had found more than 50,000 Russian-linked accounts tweeting election-related content in the 10 weeks preceding the vote. Twitter said it would email notifications to 677,775 people in the United States who followed, retweeted or liked tweets from accounts linked to propaganda efforts by the Internet Research Agency, a shadowy Russian company linked to the Kremlin.
But Mr. Ghosh and Mr. Scott said it would be a mistake to focus solely on the Russians. They argued that the tools used by the Russians could easily be applied to other misinformation campaigns. In the same way digital advertising campaigns spend relatively small sums of money to reach millions of people, any party with an interest in swaying sentiment can gain access to reams of behavioral data on the internet to target specific audiences.
Improvements in artificial intelligence for digital advertising could allow for more precise audience targeting and make the problem even worse, they said.
Fundamentally, the problem is that “disinformation campaigns and legitimate advertising campaigns are effectively indistinguishable on leading internet platforms, ” Mr. Ghosh and Mr. Scott wrote.
This month, Mr. Ghosh’s former employer, Facebook, announced changes to what its more than two billion members would see most often. The company said it would give what users’ friends and family shared priority over content from publishers and brands, with the goal of delivering a more positive experience on Facebook.
The changes were viewed as a response to the wave of criticism accusing the company of allowing its algorithms to promote misleading news and misinformation. On Friday, Facebook also said it would give priority to news sources that its users ranked as the most credible.
Mr. Ghosh said that these changes were significant and might throw off propagators of disinformation in the short term, but that they — like advertisers — would adjust. Ultimately, the underlying business model of digital advertising hasn’t changed, and it’s not clear how Facebook will handle content that advertisers pay to promote.
“I felt like Facebook was an amazing company, and in many ways I still think it is,” Mr. Ghosh said. “But something changed in me at the Javits Center.”
An earlier version of this article misidentified the center at Harvard that collaborated with New America on a report about the intersection of technology, advertising and disinformation. It was the Shorenstein Center on Media, Politics and Public Policy, not the Berkman Klein Center for Internet and Society.