A few weeks ago, YouTube demonetized conservative comedian Steven Crowder for repeatedly making derogatory comments about a homosexual Vox reporter. Unsurprisingly, this outcome made no one happy. Many on the left argued that YouTube should have removed Crowder from the platform entirely. For conservatives, the decision became just another data point indicating, anecdotally, that platforms display a political bias by removing conservative content while letting liberal content remain unfettered.

In response to this and other incidents of alleged conservative bias, Sen. Josh Hawley, R-Mo., introduced a bill this week to require large internet platforms like Facebook and Twitter to apply to the Federal Trade Commission for certification that their moderation practices do not favor one political party. Without such certification, the platforms would not be able to assert Section 230 of the Communications Decency Act (CDA 230) as a defense in litigation against them for defamatory or otherwise illegal user content. In other words, unless the platforms are certified, they would be liable for the content their users post.

Not only would this legislation drastically hurt free speech and competition in the online ecosystem, it would mean more conservative content, not less, will be removed from these websites.

CDA 230 solves what is known as the moderator’s dilemma. Under law, platforms cannot be held liable for the content that users post on those platforms unless they knew or should have known about the content. However, by engaging in good-faith moderation efforts to limit harmful content and tailor users’ experience, platforms are construed to have knowledge of any content that they have not removed. This led to a dilemma: Platforms must either moderate users’ posts and risk being held liable for their content, or refuse to moderate at all in order to avoid liability. Congress passed CDA 230 to resolve this dilemma by stating that platforms are not publishers of user-generated content, even if they take steps to moderate the content.

Sen. Hawley’s proposal effectively recreates this dilemma. Under his bill, to refuse certification, the FTC need not show that the platform actively engaged in moderation practices designed to hurt a viewpoint, but rather that its practices “disproportionately restrict[] or promote[] access to, or the availability of, information from a political party, political  candidate, or political viewpoint.” This means that, if a platform wants to moderate user content, it will have to take a rigid stance against all content so as to ensure that the FTC decides, by clear and convincing evidence, that the platform did not moderate it in a politically biased manner. In other words, legitimate speech — conservative and liberal alike — will suffer because platforms will worry that failing to remove questionable content may appear to disproportionately affect a viewpoint.

And this isn’t a one-time certification. The bill requires platforms to renew their certification every two years. With such a requirement, the threat of 230 liability remains on the horizon at all times. Therefore, even if a platform has obtained certification from the FTC, the moderation decisions it makes will continue to be largely influenced by the understanding that, every two years, the platform will need to show that its decisions did not disproportionately harm one side or the other.

If a platform fails to receive certification, the bill essentially prohibits it from asserting CDA 230 immunity for two years while it shifts its moderation practices to comply with FTC rules, whatever they may be. During this time, the platform can be sued for any defamatory content its users post. So instead of risking liability, these platforms will likely remove all content — including conservative content — that could potentially be defamatory or illegal.

If Sen. Hawley is worried about conservative voices being stifled online, his bill would just add fuel to the fire. Steven Crowder, for example, repeatedly called the Vox reporter a “lispy queer” and a “gay Latino.” Even assuming his statements did not constitute defamation or some other tort, YouTube would likely not risk a costly litigation process to defend Steven Crowder; it would just remove him from the platform entirely.

That is not to say whether the comedian should be banned from the website, nor is it to say definitively whether online platforms exhibit anti-conservative bias. It is only to say that even if platforms do display biases against certain viewpoints, this bill would only serve to further stifle these voices.