Two days, three hearings and six things we still don’t know about Russia’s election meddling on social media

After three congressional hearings in two days, there seems to be no doubt in the minds of U.S. lawmakers that Russia weaponized Facebook, Google and Twitter to spread disinformation and sow social unrest around the 2016 presidential election.

That doesn’t mean, however, that their investigation is finished.

In reality, lawmakers are really just getting started. Concerns remain about the full scope of the Kremlin’s activities, beyond the new revelations that trolls reached more than 140 million Americans on Facebook around Election Day. And lawmakers still aren’t convinced the tech industry has a full grasp on the problem.

Here are six questions that Congress still has to confront in its probe:

1. Is there any tie to Trump? It’s arguably the most significant question looming over lawmakers’ three hearings, and it’s one they didn’t — or, perhaps, couldn’t — discuss. It was only on Monday, remember, that Robert Mueller, the U.S. government’s special counsel, secured his first indictment in the wide-ranging investigation into potential Russian collusion. But it doesn’t directly implicate the president. And there was no evidence aired this week that tied any of those threads together.

Even Democrats acknowledged that fact. “Whether the Russians and Trump coordinated these efforts, we do not yet know,” said Rep. Adam Schiff, the top Democrat on the House Intelligence Committee, on Wednesday. “But it is true that the Russians mounted what could be described as an independent expenditure campaign on Trump’s behalf.”

2. What impact did this actually have on the election’s outcome? Facebook, Google and Twitter each admitted that Russian state actors used their platforms to spread disinformation during the 2016 cycle. But it isn’t clear — and, truly, it might never be clear — whether the sum total of that content affected voters’ thinking on Election Day.

To be sure, a chunk of the Kremlin-sponsored ads, stories, tweets and other content actually appeared in the weeks after U.S. voters headed to the polls. That fact led the Senate Intelligence Committee’s leader, Sen. Richard Burr, to lament media reports that have sought to attribute Trump’s victory entirely to Russia’s online campaign. And the trio of tech companies, similarly, aren’t eager to admit that they played any role in shaping the outcome of the presidential race. “We’re not well positioned to judge why any one person or an entire electorate voted as it did,” said Colin Stretch, the general counsel of Facebook, during a hearing before the Senate Judiciary Committee on Tuesday.

3. Were Russian efforts on Facebook even worse than reported? For Sen. Mark Warner, a lingering issue is whether the social giant has conducted a truly exhaustive search of the stories, posts, ads and other content on its platform. He raised it repeatedly during the Senate Intelligence Committee’s hearing on Wednesday morning.

Recall that Facebook initially identified about 470 profiles and pages tied to the Internet Research Agency, a Russia government-sponsored troll farm. The so-called IRA was responsible for getting content into perhaps as many as 126 million Americans’ feeds. But Warner said he’s as uncertain as ever if Facebook is looking hard enough for other instances of Russian disinformation outside the context of the IRA. Like every company testifying this week, though, Facebook said its internal investigation is ongoing. None of them could say they are convinced they’ve found anything.

And related, what about Instagram and other Facebook products? On Wednesday, lawmakers learned Facebook’s top-line figure — 126 million Americans saw Russian-generated organic content on the platform — did not include Instagram. There, Russian efforts reached 16 million additional accounts. Otherwise, though, the company has said little about what happened on the platform. And Facebook did not elaborate about previous comments that some Russian trolls sought to interact with users through its chat app, Messenger. (Not that it was asked, anyway.)

4. What is the real deal with bots on Twitter? In the words of Sean Edgett, the company’s acting general counsel, automated accounts comprise less than 5 percent of its users. And in the eyes of some researchers, the number might actually be closer to 15 percent. That’s a huge discrepancy, and it wasn’t lost on Sen. Richard Burr, the Senate Intelligence Committee’s leader, who asked Twitter on Wednesday to break down the numbers in written responses to the panel in the coming weeks.

5. How many people were duped into voting via Twitter or text message? One of the ways that Russia sought to stir up trouble through Twitter involved voter suppression. At least 106 since-banned accounts sent more than 700 tweets telling users they could vote using Twitter or by text message, neither of which is actually true.

But Twitter has a lot still to explain about those tweets, which appeared to target Hillary Clinton’s supporters. For one thing, how many users actually tried to vote in this manner? And, secondly, what does Twitter mean when it says there are voter suppression-related tweets from hundreds of other accounts that are “inaccessible, pending deletion”? The line appears in its testimony to the Senate Intelligence Committee on Wednesday.

6. Who is spending what, exactly? On a day when Facebook is set to report its earnings, Sen. Kamala Harris came to the hearing with securities filings for the company and its peers. Her question: How much money did Facebook, Google and Twitter make from perfectly legitimate, legal ads that ran near or alongside Russian-sponsored disinformation? None of the companies had an answer.

Sen. Jack Reed, meanwhile, pressed those companies to detail the percent of revenue devoted toward policing their platforms for such malicious content. Again, no answers, even as Facebook stressed it is hiring scores of new employees to monitor political ads. And again, they’re going to have to share more information with the lawmaker’s office after the hearing concludes.

This article originally appeared on Recode.net.