Alphabet, the parent company of Google, has dropped its principle to never use its AI in cases that are ‘likely to cause harm’ – such as for weapons and surveillance.
In a blog post, James Manyika, Google’s senior vice president, and Demis Hassabis, CEO and co-founder at Google’s DeepMind, backed the decision, saying changes reflect an evolving geopolitical landscape and a need for businesses to support national security.
With the changes in mind, the statement went on to highlight three new core AI principles: bold innovation, responsible development and deployment, and collaborative progress.
The pair also said that changes were taking place because AI as a technology had “evolved rapidly” since the rules were created in 2018.
“[AI] has moved from a niche research topic in the lab to a technology that is becoming as pervasive as mobile phones and the internet itself,” the post read.
“We believe democracies should lead in AI development, guided by core values like freedom, equality, and respect for human rights.”
“And we believe that companies, governments, and organisations sharing these values should work together to create AI that protects people, promotes global growth, and supports national security,” the blog post continued.
It’s been an ongoing challenge for businesses, AI and non-AI alike, to develop guiding principles on the use of AI when the technology is growing so rapidly and yet still in its infancy.
Google has seen backlash in the past around AI ethics, when in 2018 it was forced to halt AI work with the US Pentagon – America’s defense headquarters – after employees pushed back through resignations and petitioning.
Alphabet’s blog post was published ahead of its earnings call, which showed weaker results than market expectations. Google is one of the tech companies impacted by the advent of Chinese ChatGPT rival DeepSeek, which sent shockwaves through US tech markets.
For Alphabet, this dip in stock value occurred despite a 10 percent increase in revenue from digital advertising. The tech giant also said it plans to spend $75 billion on AI projects in 2025 – 29 percent more than economic analysts expected.
Google has launched several AI products over the years, including Google Assistant (2016), Google Translate’s Neural Machine Translation (2016), Google Lens (2017), Google Duplex (2018), Bard (2023), and Gemini AI models (2023) – with the later models coming out of Google’s AI division DeepMind.