I find this latest update from Google & Alphabet quite interesting if not a little unsettling. The famous motto, later downgraded to mantra of, ‘Do No Evil’ has been removed in a recent update to their AI Ethical Guidelines released in 2018, updated in 2025.
Back in 2018 the company Google/Alphabet partnered with the Pentagon – US Military Defence with project, Maven. So, no, not the build tool you may be familiar with. Maven implemented Google artificial intelligence technology tensorflow. Initially it was used for TOIL reduction and routine tasks for efficiency. This was grounded in non-combat operations and not offering weapon capabilities. For example analysing rust on battleships and helping to maintain its airforce fleet.
“While we are not developing AI for use in weapons,” – “we will continue our work with governments and the military in many other areas.”
Sundar Pichai
There are serious ethical questions when a publicy owned and publicly consumed mega corporation funded largely ostensibly by consumer activity works with the military in the first place. Millions of devices, handsets, chromebooks, apps and connected technology use google. This would have serious implications concerning surveillance especially.
Thousands of Google staff protested in 2018 when Google explicitly announced it engaged in the project. At that time Google made clear it “Would Not” use the AI for weapons, or surveillance projects. Sundar Pichai commented, “While we are not developing AI for use in weapons,” – “we will continue our work with governments and the military in many other areas.”
So Google/Alphabet’s latest update u-turns the previous statement. Between 2018 – 2021 and perhaps more recently Google/Alphabet had been true to that word. Relatively recently it won contracts including a CIA Cloud Contract and Co-Win with Amazon AWS and Google (GCP) to supply cloud services to the Israeli Defence Force. Many have cited this is why its important to get Laws and Regulations because Law cannot be so easily abrogated.
This update does coincide with lower than forecasted earnings for the tech giant. I must honestly say that meta search is not what it used to be and the content I search for on google is lower and lower down the page fold so that getting number. one SERP ranking definitely does not mean as much as it used to. I wish Silicon Valley would innovate more and google would focus on bringing new products to market, rather than selling up all its screen real estate.
Clearly the ethical trope continues. Essentially, where money is involved there is influence, so to what extent could the the military influence Google, especially their roadmap for AI. Given its slight loss in Cloud dominance to Microsoft/Azure and Amazon AWS. Its not an impossibility.
Recently Biden attested in his farewell address: 15th Jan 2015
J Biden, Jan 2015
Today, an oligarchy is taking shape in America of extreme wealth, power, and influence that literally threatens our entire democracy, our basic rights and freedoms, and a fair shot for everyone to get ahead.
I also think the words of Thedore Roosevelt are of value to as is the whole speech:
Theodore Roosevelt, 1910
At many stages in the advance of humanity, this conflict between the men who possess more than they have earned and the men who have earned more than they possess is the central condition of progress. In our day it appears as the struggle of freemen to gain and hold the right of self-government as against the special interests, who twist the methods of free government into machinery for defeating the popular will. At every stage, and under all circumstances, the essence of the struggle is to equalize opportunity, destroy privilege, and give to the life and citizenship of every individual the highest possible value both to himself and to the commonwealth. …
Now here’s the rub.
As citizens of sovereign nations we collectively cede governance to the government to act in the best interests of the nation i.e. the people. So there is a counter argument. Namely, it would be in the best interests of the people to allow the millitary to understand, synthesise and deploy any offensive capability that would give an advantage to the protection of the nation state. That sounds totally reasonable and I agree in principle. However there are practical security concerns that in creating any offensive technology opens a backdoor where it could possibly fall into the wrong hands. All it would take is a security slip, whether by a bad actor, foreign influence infiltration, or the phsyical hardware like a drone or hdd literally falling into the wrong hands. If such were to happen the potential affects on the CIA triad would be immense and far reaching.
I would not suggest that Google/Alphabet is going full Skynet meets StarCraft. But in decades it may eventually get down to that. One to watch. Google you’ve done an immense amount of good over the years, even Google Wave was interesting for a time 🙂
Collectively as consumers, technologists and innovators there is power to hold organisations no matter how large to account. I just think it’s a grey area when a publicly advertised and public identity b2b, b2c goes into providing weapon based capabilities and surveillance especially in the Wake of Facebook/Meta and Cambridge Analytica. I personally don’t know why more has not been said and why people are not talking more about this?
Probably too busy waiting for Jurassic Park Rebirth or The Last of Us, Season 2.
- Links:
Update from Google 2025 – James Manyika & Demis Hassabis,
https://blog.google/technology/ai/responsible-ai-2024-report-ongoing-work/ - Biden Farewell Address
https://bidenwhitehouse.archives.gov/briefing-room/speeches-remarks/2025/01/15/remarks-by-president-biden-in-a-farewell-address-to-the-nation/ - Barack Obama
https://medium.com/obama-white-house/president-obamas-farewell-address-d6f45155d245 - Theodore Roosevelt
https://www.americanyawp.com/reader/20-the-progressive-era/theodore-roosevelt-on-the-new-nationalism-1910/ - BBC
https://www.bbc.co.uk/news/articles/cy081nqx2zjo - The Guardian.
https://www.theguardian.com/technology/2025/feb/05/google-owner-drops-promise-not-to-use-ai-for-weapons - Google Responsible AI
https://blog.google/technology/ai/responsible-ai-2024-report-ongoing-work/ - Google AI Principles
https://ai.google/responsibility/principles/
- Other links
- https://www.wired.com/story/google-wont-renew-controversial-pentagon-ai-project/
- https://gizmodo.com/google-removes-nearly-all-mentions-of-dont-be-evil-from-1826153393
- https://en.wikipedia.org/wiki/Project_Maven
- https://www.defense.gov/News/News-Stories/Article/Article/1254719/project-maven-to-deploy-computer-algorithms-to-war-zone-by-years-end/
- https://www.independent.co.uk/tech/google-artificial-intelligence-project-maven-ai-weaponise-military-protest-a8352751.html
