Filter pornography automatically in your social product
Request a Demo
Send us your details in the form below, and we'll be in touch within 24 hrs.
Have you considered how you will deal with pornographic images in your social product?
When you allow community members to share User Generated Content (UGC), you can expect about 1% of posted images to contain pornography, regardless of community guidelines. And when left unmoderated, we have seen that number grow to as much as 10%.
You could pre-moderate all images, but as your community grows that solution will quickly become unrealistic and expensive. And why force your moderators to review and reject images that a computer can find in a fraction of the time?
When unwanted sexually explicit images appear in your product, your brand is affected negatively. To protect your community and your brand reputation, it’s crucial that you use the most advanced technology available to prevent the sharing of explicit nudity on your platform.
Leverage Powerful Image Recognition
Community Sift is your cutting-edge, cost effective solution. We have partnered with the world’s leading image analysis filter and combined it with our patent-pending user reputation system and powerful text classifier. Our detection rate in identifying pornography is over 99%.
That means you can rest easy knowing that the newest technology — deep learning and a robust neural network — is watching over your community, day and night.
And we are continually updating and improving our technology by training our Artificial Intelligence (AI) on new datasets. Image analysis is a growing technology that never rests.
Do you host of any of the following in your social product?
- Profile pics
- Photo contest entries
If so, our image analysis will save you hours of wasted time and effort that it takes to manually review uploaded images, all while providing unprecedented levels of accuracy.
What are the benefits of using Community Sift to moderate images in your product?
Filter Pornography in Real Time
Your community never sees it. When you protect users from explicit imagery, you protect your brand reputation.
We’ve mapped the probability that an image is pornography to our Community Sift risk levels. So you choose your tolerance level, based on your unique audience.
Mature audiences will likely filter obvious pornography and depend on user reports to catch blurry or unclear images. Child-directed products will be stricter, filtering images with a lower probability of being pornographic.
Cut Moderation Costs
Let our deep learning algorithm approve or reject images automatically, freeing up your moderation team to focus on what matters — reviewing user reports, making tough decisions, and fostering community engagement.
Our detection rate is over 99% in identifying pornography. Accuracy is crucial when filtering pornographic images in your product. You should never worry that you’ll compromise the safety of your community.
Automated Image Filtering
Using a simple API call, you can moderate any image and access our patent-pending reputation and risk level assessment system.
- User submits a new image. You call our API with the image URL, user ID, and tolerance level.
- Our deep learning algorithm scans the image and compares it to a massive database of pornographic images. It returns a percentage based on the probability that the image is pornographic.
- We compare that percentage to your tolerance level. Does it pass or fail?
- If it’s somewhere in between, we review the user reputation. Do they often post images that fail your tolerance level? If so, we return a False response.
- If False, the image is rejected.
- If True, the image is automatically approved or added to a queue for human moderation.
Users can also report approved images, and the cycle starts again. If the system returns a questionable probability percentage, we recommend that you automatically remove the image.
If the response is True, we recommend that you see if multiple reports are filed before taking action.
Running a child-directed or family-friendly product? The process is the same, but with an added safeguard.
We recommend that your team pre-moderates any image that the system has labeled questionable, especially those submitted by new users. Consider making image uploads an “earned” feature that only users with the best reputations can access.
We have settings that work for every audience. And our team of industry experts will help guide you every step of the way.
Keep Explicit Imagery Out of Your Community
When you protect your community from nudity and pornography, you protect your business and your reputation. As your community grows and your userbase posts more content, you’ll need an automated solution to keep them safe from explicit images.
Our combination of automation with human teams reviewing user reports is a powerful tool in the hands of social products. Give your community — and your business — peace of mind and protect them from explicit imagery with our advanced pornography filter.
We would love to start a conversation about how we can help you thrive. Contact us today for a free demo, and let’s talk.