Artificial intelligence companies join Thorn initiative to protect children online – Internet

Thorn and All Tech is Human are leading an initiative to stop AI tools from being used to exploit children or generate pedopornographic material. More than a dozen artificial intelligence companies, including OpenAI, Microsoft, Google and Meta have committed to complying with generative AI principles to promote child safety online, for example by integrating safety by design.

Thorn is a child safety group and All Tech Is Human is a nonprofit focused on responsible technology.

The aim of this initiative is to prevent the creation of AI-generated child sexual abuse material (AIG-CSAM) and other sexual harm against children and remove them from social media platforms and search engines.

More than 104 million files of suspected child sexual abuse material were reported in the US in 2023 alone, Thorn says. In Portugal, last year, more victims of online scams sought support from the Safe Internet Line. The number of victims of image-based sexual violence and reports of sexual crimes against children have also increased.

On Tuesday, Thorn and All Tech Is Human released a new document titled “Safety by Design for Generative AI: Preventing Child Sexual Abuse”which defines strategies and presents recommendations for companies that build AI tools, search engines, social media platforms, hosting companies, and developers to take steps to prevent generative AI from being used to harm children.

One of the recommendations, for example, asks companies to carefully choose the datasets used to train AI models. It also asks them to avoid data that only contains instances of What is Child Sexual Abuse Material (CSAM)but also adult sexual content, due to generative AI’s propensity to combine the two concepts.

Safety by Design

credits: Thorn

” data-title=”Safety by Design – Artificial intelligence companies join Thorn initiative to protect children online – SAPO Tek”>

Thorn also asks social media and search engines to remove links to websites and applications that allow the viewing of nude images of children.

According to Thorn, Some companies have already agreed to separate images, videos and audio involving children from datasets containing adult content, to prevent your models from combining the two. Others also add watermarks to identify AI-generated content, but the method is not foolproof: watermarks and metadata can be easily removed.

Protecting children online, due to the growth of generative AI, has been a recurring theme in recent months. The UK regulator for online safety has plans to use AI and other automated tools to detect and remove illegal content online and protect children from accessing it.

In Canada, a legislative proposal wants to force social networks to remove offensive content within 24 hours, following Europe’s footsteps in tightening laws against online sexual violence, especially aimed at children.

In February, the new rules of the Digital Services Regulation (DAS) began to be applied. With this document, the EU aims to make online browsing safer, fairer and more transparent for users, applying it to all online platforms.

From a reactive perspective, authorities are also paying attention. This month, a man was arrested for obtaining pornography images of minors on chat applications.