Automated Content Advisories

The web has a content problem. It's pervasive. It's persistent. And it centers on persuasive langauge. We have to do better, and this starts with a duty to warn.

WARNING

This content appears to contain a high level of rhetoric, and may be trying to persuade, manipulate, or otherwise influence you.

User discretion is strongly advised.

The world is very different now. For man holds in his mortal hands the power to abolish all forms of human poverty and all forms of human life. And yet the same revolutionary beliefs for which our forebears fought are still at issue around the globe -- the belief that the rights of man come not from the generosity of the state, but from the hand of God.

We dare not forget today that we are the heirs of that first revolution. Let the word go forth from this time and place, to friend and foe alike, that the torch has been passed to a new generation of Americans -- born in this century, tempered by war, disciplined by a hard and bitter peace, proud of our ancient heritage, and unwilling to witness or permit the slow undoing of those human rights to which this nation has always been committed, and to which we are committed today at home and around the world.

Let every nation know, whether it wishes us well or ill, that we shall pay any price, bear any burden, meet any hardship, support any friend, oppose any foe, to assure the survival and the success of liberty.

This much we pledge -- and more.

-- J.F.K.

Improve User Experience

Protect Your Advertisers

We put advisories on movies, television, prescription drugs, and tobacco. Why can't we put them on internet content, too?

Auto-Generate Advisories For Content

Rhetoric

Our flexible API works with text, audio, and video, meaning you can get advisories for a wide range of products and services, such as

Search Engines

Content Platforms

Social Media

And LLM Output

just to name a few.

If your content has 100+ words, odds are we can assess its rhetoric

"It's not what you said, it's how you said it."

-- Uttered by every parent ever, at one point or another. It's also how our rhetoric models work.

Importantly, our rhetoric models generally err on the side of caution, because people deserve the benefit of doubt. And they definitely shouldn't be subjected to the biases of word embeddings. So we don't use them. Or neural networksinfo for that matter. This helps keep our rhetoric models agnostic self_improvement

Results are based off 1250 samples. A detailed whitepaper is coming soon(ish).

Themes

But sometimes you want more than a simple advisory. Our Theme API can help identify themes in text, audio, and video content, meaning you can generate more nuanced messaging.

WARNING

This content appears to contain a high level of rhetoric, and may be trying to persuade, manipulate, or otherwise influence you.

Themes

• Panic
• Obsession
• Fairness

User discretion is strongly advised

Additional Features

Hate-Speech

People should not be subjected to the hateful nonsense that many internet denizens sadly throw around. Our hate-speech and insult detector can help you expand your advisories beyond rhetoric, giving you the power to provide users with more info about content.

Profanity

Naughty words? Not on your watch! Our profanity detection tool works with text, audio, and video, and can help you build kid-friendly versions of your service.

Nudity

There is a time and place for nude and/or pornographic materials. Give your users a heads-up about these materials so they aren't caught off-guard. Or simply filter nudity out of your app. You can do both with our nudity detection features.

Data & Privacy

  • We store your submitted data (text/audio/video content) for up to 30 days, at which point it is deleted
  • We only store and use this data for debugging purposes
  • We do not use this data to train ML models
  • We do not make this data available to anyone outside the company
  • We retain a small set of personal information needed for basic site functionality (e.g., emails)
  • Some of this personal information is made available to third-party integrations (e.g., payment processors)