Article · AI translation
IT Home reported on November 3 that, according to Bloomberg, Neil Sethi had for years posted Wall Street research reports on X, along with charts and stock reference information, gradually building a following of nearly 30,000. He saw it as a mutually beneficial practice: he built his personal profile, while the banks’ research reports gained more exposure through likes and reposts.
But this summer, everything changed. Bank of America accused him of posting its material without permission, and his account, @neilksethi, was subsequently suspended. Sethi runs a small investment firm in Upper Arlington, Ohio. He promised never to use the relevant research material again, but still has not regained the account he had operated for 10 years.
“That account helped me connect with many people I otherwise would never have crossed paths with,” said Sethi, 54. “It’s really frustrating to have an account suspended just like that, without even a chance to understand what happened or fix it.”
Sethi’s predicament is playing out repeatedly across a large online community commonly known as FinTwit, or financial Twitter, a name derived from X’s former name. Many industry observers believe it reflects a shift in attitude among several leading Wall Street firms: to protect their business models, they have begun taking tough measures to control how widely their research reports circulate.
Over the past 20 years, strict regulations governing charges for bank research, combined with the rise of index funds, have dealt a severe blow to research departments at institutions large and small. Now, with artificial intelligence advancing rapidly, a firm’s exclusive analysis can easily become widely available content once a report reaches the internet, further heightening banks’ anxiety.
“Pirated content has been around for a long time. It’s just that AI agents and social platforms now make these research reports especially easy to find, so this kind of activity is becoming more common,” said Ben Mok, vice president of brand and content protection at Corsearch, which files complaints on behalf of Bank of America. He declined to discuss the details of Bank of America’s complaints.
“Cybersecurity teams are starting to pay attention to this now, too.”
Of particular concern is that web-scraping models directly ingest posts containing bank research reports to improve their own outputs, often without citing the source. Banks worry that, as AI chatbots and agents combine information from many reports to generate answers, bank research will lose value, or at least that their ability to charge for analysts’ views will be greatly weakened.
Banks have been tight-lipped about the matter. Bank of America declined to comment on this crackdown, as did Morgan Stanley, which has also filed complaints against X accounts in the FinTwit community.
Nick Carcaterra, a spokesperson for Goldman Sachs Group, said the bank does not allow outsiders to post its research reports on social platforms, but did not say whether it had taken steps to enforce the ban.
Other Wall Street institutions, including JPMorgan Chase, Citigroup and Toronto-Dominion Bank, were also unwilling to say whether they had taken similar action. Royal Bank of Canada and X did not respond to requests for comment.
Complaint documents seen by Bloomberg show that at least a dozen FinTwit accounts, with follower counts ranging from hundreds to tens of thousands, have received Digital Millennium Copyright Act infringement notices from Bank of America or Morgan Stanley.
Ryan Detrick, chief market strategist at Carson Group Holdings LLC, received his first notice from Morgan Stanley on September 2, and at one point could not log in to his X account, which has more than 225,000 followers. Financial writer Mike Zaccardi, who has more than 80,000 followers, said he began receiving copyright notices on behalf of Bank of America about a year ago, with a surge of them in recent months. Sethi and Zaccardi also received copyright complaints forwarded by LinkedIn after posting bank research reports there.
“So much material is electronic now, and generative AI is far better at retrieving information than ordinary people,” said Michael Eastwood, head of Americas equity research at Jefferies. “That makes banks’ content easier to scrape.”
The issue has long since spread beyond Wall Street. From digital media to e-commerce, industries are working to protect their content from being scraped by bots. According to cybersecurity company Cloudflare, crawler traffic now accounts for 65% of all web traffic.
In one widely watched case, The New York Times and the Chicago Tribune sued Perplexity AI late last year, accusing the startup of copying and distributing their content without authorization.
Banks’ research departments already have ample reason to be wary. Over the past 10 years, their standing in the industry has continued to decline, while client spending has kept shrinking.
IT Home notes that regulations such as the U.S. Markets in Financial Instruments Directive II require banks to price research separately from trading commissions, directly shrinking research budgets. The widespread adoption of ETFs has fueled a boom in passive investing, with market participants more inclined to trade broad themes, reducing the importance of stock-picking analysts.
Meanwhile, hedge funds and other clients that once paid reliably have been building in-house research teams. According to research firm Substantive Research, clients’ spending on equity research has fallen by half since 2018.
Rebecca Tushnet, a Harvard Law School professor specializing in the First Amendment, said factual information such as market data is not itself protected by copyright, but the arrangement and presentation of that information are.
“If you copy it exactly as it is, there is indeed a reasonable basis for a copyright infringement claim.”
Sethi himself acknowledged that he had not sought Bank of America’s permission to use the research report, which he obtained through an account he had opened with the bank’s wealth management division. Like many in the FinTwit community, he assumed the banks tacitly allowed this kind of sharing.
Zaccardi thought the same way. At first, he did not take the complaints seriously because he saw Bank of America research reports all over X.
“Even now, I can see plenty of accounts that keep posting Bank of America material, and those accounts have far more followers than I do.”
Despite the warnings and the risk of suspension, Sethi and Zaccardi still regularly repost bank research reports, including Bank of America material. In this AI era, both acknowledge using AI to alter charts so they look different from the originals, which is perhaps unsurprising.
The threat of having an account suspended at any time still hangs over them. Sethi puts it this way: his new X account has only about 9,300 followers, far fewer than his previous one. “It’s really hard to rebuild those connections,” he said.