0 
0 Elon Musk has said he won’t move forward with his $44 billion Twitter acquisition until he has more details about fake accounts on the platform, but he met with the company’s top brass for three days to discuss its business before he publicly announced his bid, according to a new securities filing.
The filing doesn’t specify what was discussed in those meetings or if Musk brought up his concerns about the bots during them.
In late March and early April, Musk held discussions with Twitter co-founder and former CEO Jack Dorsey, current CEO Parag Agrawal, Twitter board chair Bret Taylor and board member Egon Durban, among other company executives.
The meetings took place after Musk invested in Twitter and before his announcement on April 14, that he was making a formal bid for the company. Twitter’s board eventually agreed to sell to Musk for $44 billion late last month, but the future of the deal is uncertain as Musk has said the acquisition will not move ahead until he has a clearer sense of the number of fake accounts on the platform.
Investors have been dumping Twitter shares on concerns that Musk is going to abandon his agreement to purchase Twitter for the agreed upon price of $54.20 a share. Twitter’s stock has given up all its gains since Elon Musk first disclosed his 9% stake in the company last month. Shares of Twitter were up 3% on Tuesday to $38.54, below the $39.31 closing price on April 1, the last trading session before Musk revealed his minority ownership in Twitter.
On Tuesday, Musk doubled down on his conviction that the Twitter deal “cannot move forward” until the company can prove that bots make up less than 5% of users on the platform. Bots are automated accounts that can be useful or nefarious. Neither Musk nor Twitter have said exactly how they define bots or fake accounts.
“My offer was based on Twitter’s SEC filings being accurate,” Musk tweeted early Tuesday morning. “Yesterday, Twitter’s CEO publicly refused to show proof of <5%. This deal cannot move forward until he does.”
Since its IPO in 2013, the company has estimated in financial filings that fake accounts or spam accounts made up less than 5% of monthly users. In its annual report for 2018, Twitter added that the number also applies to its monetizable daily active users (mDAUs).
The company, which had 229 million mDAUs as of last quarter, says “the actual number of false or spam accounts could be higher than we have estimated.”
In a series of tweets on Monday, Agrawal broke down how Twitter determines what percentage of accounts are fake on the platform. He said Twitter can’t publicly disclose specific details of the process because the company relies in part on private user information.
Musk replied to one of Agrawal’s tweets with a smiling poop emoji, then said in a separate tweet: “So how do advertisers know what they’re getting for their money? This is fundamental to the financial health of Twitter.”
Musk further elaborated on his thoughts about Twitter’s spam problem on Tuesday at a summit hosted by Chamath Palihapitiya, Jason Calacanis, David Sacks and David Friedberg for their “All-In” podcast.
“It seems beyond reasonable for Twitter to claim that the number of real, unique humans that you see making comments on a daily basis on Twitter is above 95%,” Musk remarked. “That is what they’re claiming. Does anyone have that experience? I mean, really?”
Musk hasn’t provided any evidence that Twitter’s calculations are unreliable. Chris Kelly, Facebook’s former chief privacy officer and general counsel, told CNBC in an interview on Tuesday that Twitter’s bot estimates are “pretty well vetted.”
On Tuesday, Musk said he estimated that around 20% of the accounts on Twitter are fake or spam, and he said he’s concerned that the number could be even higher.
“Obviously, there could be and there should be challenges from outside sometimes, but Elon doesn’t seem to have any evidence,” Kelly said. “…Whereas Parag and the Twitter team have presented a lot of evidence about how they do this. Barring him coming up with real evidence here his assertions are just assertions.”
Twitter has previously faced criticism over the accuracy of its user metrics. Last September, Twitter said it agreed to pay $809.5 million to settle a class-action lawsuit brought in 2016 by shareholders who argued it artificially inflated its stock price by misleading them about user engagement.
Musk has his own ideas on how to quantify the number of fake, spam and duplicate accounts on Twitter. Last week, Musk said in a tweet he would review “a random sample of 100 followers of @twitter.”
He added later: “Ignore first 1000 followers, then pick every 10th. I’m open to better ideas.”
Experts in social media, disinformation and statistical analysis told CNBC that Musk’s approach won’t work and should not serve as “due diligence” for making a $44 billion acquisition.
— CNBC’s Lora Kolodny contributed to this story.
WATCH: Elon Musk doesn’t seem to have evidence for his bot claims, says former Facebook chief privacy officer
