With 95 million photos and videos uploaded to Instagram every day, manual content moderation is impossible. Inappropriate content must be reported for moderation today, and the psychological damage it can do to human moderators is well-documented.

AI has long been herald as a means to automate the moderation process, but today it still requires a human to make a final decision. Instagram has decided to follow the old adage of ‘prevention is better than a cure’ with its new AI system.

Instagram already uses such a system for negative comments, providing users with the opportunity to undo the message. Instagram claims in earlier tests they’ve found “it encourages some people to undo their comment and share something less hurtful once they have had a chance to reflect.”

Now, prior to any potentially harmful content being shared, the platform will intervene to ask “are you sure you want to post this?” to really make the user think about what they’re doing.

Instagram is also helping to stop bullies getting the attention they seek from their victims.

“We’ve heard from young people in our community that they’re reluctant to block, unfollow, or report their bully because it could escalate the situation, especially if they interact with their bully in real life,” says Instagram head Adam Mosseri.

Users flagged as bullies will be made invisible to the victim. Furthermore, the victim’s current online status will be made unknown to the bully.

“Restricted people won’t be able to see when you’re active on Instagram or when you’ve read their direct messages,” adds Mosseri.

A study last year found that around 59 percent of US teens have suffered from bullying and harassment online. Over half of the respondents said that social media sites, police, teachers, and others are not doing enough to counter the issue.

The new Instagram AI intervention system will begin rolling out to English speakers first before it’s made available globally.

deepgeniusai.com/">AI & Big Data Expo events with upcoming shows in Silicon Valley, London, and Amsterdam to learn more. Co-located with the IoT Tech Expo, , & .

The post Instagram uses ‘AI intervention’ to help counter bullying appeared first on AI News.

The social media giant acquired Instagram for $1 billion in 2012. At the time, some criticised its value with its 30 million users. Now, however, the platform has over 800 million active monthly users.

Facebook trained a machine learning algorithm to categorise images by itself using over 3.5 billion posted on Instagram. The algorithm took into account the 17,000 hashtags used for the images.

Speaking during Facebook’s F8 developer conference, CTO Mike Schroeper said:

“The biggest limiting factor to making progress in computer vision — as in many fields of AI — is that we rely almost entirely on hand-labelled, human-curated, data sets.

If a person hasn’t spent the time to label something specific in an image, even the most advanced computer vision systems won’t be able to detect it at runtime because it hasn’t seen it in the training set.

We built some breakthrough technology that takes publicly available hashtagged images at an unprecedented scale. We have trained on 3.5 billion training images using a public set of images without any human curated images in that data set.”

That image set is 10x the size of the 300 million Google uses to train their algorithms, which the company disclosed in a paper (PDF) released in August 2017.

Srinivas Narayanan, Engineering Director at Facebook, added:

“We have now created the world’s best computer vision system. It achieves the highest score ever of 84.5% accuracy on ImageNet — a dataset widely used for benchmarking.”

Back in August last year, AI News reported that IBM had set a record for image recognition with a 33.8 percent accuracy rate after analysing 7.5 million pictures over seven hours. Facebook may have just set the new standard.

What are your thoughts on Facebook’s computer vision achievement?

The post Facebook trained its AI using 3.5 billion Instagram images appeared first on AI News.