Google pubblica il Safety Report 2023: l’IA ha bloccato 5,5 miliardi di annunci thumbnail

Google publishes the 2023 Safety Report: AI blocked 5.5 billion ads

The fact that artificial intelligence is an increasingly topical topic is also well expressed by the number of articles we dedicate to the topic.

An increasingly complex and multifaceted topic, especially when generative AI has appeared in our lives in a disruptive way. Bringing enthusiasm and apprehension, showing more or less virtuous uses and an increasingly burning desire to give limits and rules to this instrument.

The areas of use of artificial intelligence are also very varied. Let's think about medicine: according to a recent study, AI would be able to predict Alzheimer's 7 years in advance.

Or the fact that AI can act as a sentry to ensure safe browsing for web users. This is what Google reminds us, which published the 2023 Safety Report, which shows the results of artificial intelligence in relation to advertisers' messages.

google 25 years of online search 1google 25 years of online search 1

Il Google Safety Report 2023

A long note from Google on the 2023 Safety Report was published on Wednesday 27 March on the company blog.

The report is introduced as follows: “Billions of people around the world rely on Google products to provide relevant and reliable information, including ads. That's why we have thousands of people working around the clock to safeguard the digital advertising ecosystem.

Today we release our annual Ad Safety Report to share the progress we've made in enforcing our policies for advertisers and publishers, and to hold ourselves accountable in our work to maintain a healthy, advertising-supported Internet.”

Artificial intelligence and policy-violating ads

The 2023 Google Safety Report then mentions the impact that generative AI has had on the world in general. And specifically in making the rules apply to those who publish ads.

In 2023, control teams took advantage of Large Language Models (LLM), which have become much more refined than in the past. The example is taken of unreliable financial statements: to be clear, ads promoting get-rich-quick schemes. Fraudulent techniques are constantly evolving, as is the financial market itself. Yet LLMs are able to quickly recognize new trends in financial services, distinguishing legitimate activity from scam attempts.

The Gemini impact

Google, in the 2023 Safety Report, then explains the great benefits resulting from when Gemini, its most advanced artificial intelligence, was integrated “into our security and enforcement initiatives”.

However, if artificial intelligence is refined, the scammers' techniques are also refined. Let's think about deepfake, used massively especially between the end of 2023 and the beginning of 2024, says the report, to deceive users.

However, also thanks to AI, 206.5 million advertisements were blocked or removed for violating the misrepresentation policy.

The numbers for 2023

Google, in the Safety Report, informs us that In 2023, more than 5.5 billion ads were blocked or removed (a slight increase compared to the previous year) and 12.7 million advertiser accounts (almost double compared to 2022).

The publication of ads on over 2.1 billion publisher pages was then blocked or limited, a slight increase compared to 2022. And, in the operation to combat widespread or serious violations, measures were taken against over 395,000 publisher sites, “a sharp increase compared to 2022”.

Safety Report 2023: how much AI has influenced

Google announces that in 2023 as many as 90% of advert checks were done with the help of LLMs. Who are capable of quickly reviewing and interpreting large quantities of content, with advanced reasoning skills, precisely to quickly spot fake ads, fraudulent services and scams.

Explains Duncan Lennox, vice president and general manager of Ads Privacy and Safety within Google: “Of course, any advertiser or publisher can still appeal an enforcement action if they believe we have done something wrong. Our teams will review it and, in cases where we find errors, use it to improve our systems.”

Don't miss this week on techgameworld.com

✒️ Our unmissable Caffellattech newsletter! Sign up here
🎧 But did you know that Fjona also has her own newsletter?! Sign up to SuggeriPODCAST!
📺 You can also find Fjona on RAI Play con Touch – Fingerprint!
💌 Let's solve your heart problems with B1NARY
🎧 Listen to our unmissable podcast Tech life
💸And you can find some interesting offers on Telegram!

Walker Ronnie is a tech writer who keeps you informed on the latest developments in the world of technology. With a keen interest in all things tech-related, Walker shares insights and updates on new gadgets, innovative advancements, and digital trends. Stay connected with Walker to stay ahead in the ever-evolving world of technology.