Google uses its artificial intelligence systems to help people access essential information while avoiding potentially shocking or harmful content, so they can stay safe, both online and offline. Safe Search is a feature in Google Search and Google Images that acts as an automatic filter for pornography and potentially offensive and inappropriate content.
Using artificial intelligence like Google would make it safer
Although Safe Search is not 100% reliable , it helps to filter explicit content in Google search results for all your image, video and website queries. In one of the latest blogs from Google , it was explained how the search engine uses MUM and BERT for more secure search results.
- Google is using MUM to better detect when a query indicates a user is in crisis and will start rolling out these improvements within a few weeks.
- Google uses BERT to improve its understanding when a user searches for explicit content.
- Using BERT in this way has reduced “unexpected shock results” for search engines by 30% in the past year, according to Google .

Better understanding
MUM can better understand the intent behind people’s queries, which helps more reliably show relevant information that’s useful at the right time.
“MUM not only understands language, but also generates it. It is trained in 75 different languages and many different tasks at once, which allows it to develop a more comprehensive understanding of information and knowledge about the world than previous models ” said Google Associate Pandu Nayak in a blog post.
“...people in personal crises search in various ways and it is not always obvious to us what their needs are. And if we can’t recognize that accurately, we can’t code our systems to display the most useful search results.” , says Nayak .
Using machine learning to improve language understanding helps Google more accurately detect when search results should contain phone numbers of relevant crisis lines, for example.
No unpleasant surprises
Unexpected search results can often be an unpleasant experience. That is why it is essential that Google is able to better understand the intent of each searcher so that the results presented to them are in line with their expectations.

Safe Search mode allows search engines to filter out explicit results. However, there are times when this is exactly what a person may be looking for.
“BERT has improved our understanding of whether searches are really looking for explicit content, helping us greatly reduce the chances of encountering surprising search results ,” Naiak wrote.
” Over the past year, using BERT in this way has reduced ‘unexpected, shocking’ results by 30% ,” he revealed.
According to Naiak , BERT was ” particularly effective at reducing explicit content for searches related to ethnicity, sexual orientation, and gender .”
Reducing useless content
Google uses artificial intelligence to reduce spam emails and useless results on various sites.
And in the coming months, MUM will try to increase these security measures even where there is very little data. This is possible because, as Naiak explained, ” When we train a MUM model to do a task — like classifying the nature of a query — it learns to do it in all the languages it knows.”
Google has assured users that these latest changes have been and will continue to be rigorously tested, including evaluation by manual search raters.
For more news and interesting stories from digital marketing , visit our blog page or follow our Instagram profile.
Made by Nebojša Radovanović – SEO Expert @Digitizer
