Dr. Luiza Jarovsky is the co-founder of the AI, Tech & Privacy Academy and one of the world's most influential voices in AI governance.
Her newsletter, with over 89,000 subscribers, is a leading publication on the legal, ethical, and social challenges of AI, and her pioneering AI Governance Training, with more than 1,400 participants to date, empowers the next generation of leaders in the field.
She is frequently cited by worldwide media outlets for her insights on AI policy and regulation.
Dr. Jarovsky was named a Top Voice in AI by LinkedIn, received a Westin Scholar Award from the IAPP, and was honored with the President's Scholarship for Excellence in Science and Innovation by the President of Israel for her Ph.D. research.
She holds a law degree from the University of São Paulo and earned her Master’s and Ph.D. from Tel Aviv University. A speaker of English, Portuguese, French, Spanish, Italian, German, and Hebrew, she is also a proud mother of three.

Relevant links and contact form: www.luizajarovsky.com

Luiza Jarovsky

Luiza Jarovsky is a lawyer, co-founder of Implement Privacy - where she leads global training programs in privacy & AI - and the author of The Privacy Whisperer newsletter and podcast, a leading publication in the field. 80,000+ people follow her work, and hundreds of privacy & tech leaders have attended her masterclasses.

In 2023, Luiza was named by LinkedIn a Top Voice in AI; in 2021, she received a Westin Scholar Award from the IAPP, and in 2020, she was awarded by the President of Israel the “President's Scholarship for Excellence in Science and Innovation” for her Ph.D. research.

Before focusing on privacy & AI, she worked as a corporate lawyer, advising entrepreneurs. She published three fiction books and edited the book “Law for Entrepreneurs” (in Portuguese).

Originally from Brazil, Luiza graduated in Law from the University of São Paulo, completed her Master's at Tel Aviv University, and is doing her Ph.D. there as well. She speaks English, Portuguese, French, Spanish, Italian, German, and Hebrew and is a proud mother of 3.

For all relevant links and to get in touch: www.luizajarovsky.com

Co-founder of the AI, Tech & Privacy Academy (1,400+ participants), Author of Luiza’s Newsletter (89,000+ subscribers), Mother of 3

Andrew Midura

🚨 This paper will transform how copyright law is applied to AI, and the paragraph below will be cited in every single lawsuit:

"LLMs memorize portions of their training data, these memorized data are encoded in the model’s weights, and, as we show here, it can be feasible to extract large quantities of in-copyright training data from production LLMs."

Download the paper below.

Luiza Jarovsky, PhD

🚨 BREAKING: Italy did it again! The Italian Data Protection Authority issued a warning reminding everyone, including xAI developers, that undressing people without their consent can be ILLEGAL and lead to criminal offenses:

The authority wrote that people who use AI to process personal data (including voice or image data) must comply with the law, in particular the GDPR, including by first providing accurate and transparent information to the people affected.

The authority writes that failing to do so might violate the GDPR, especially Articles 5.1.a, 6, and 9, and that sanctions may apply. There might also be criminal offenses involved.

They also wrote that services like Grok make it extremely easy to misuse images and voices of third parties, without any legal basis, and that they are working with the Irish data protection authority and 'reserve the right to take further action.'

When dealing with AI, The Italian Data Protection Authority has often been the most proactive European authority, making it clear what is wrong, unethical, and illegal, and warning the parties involved.

If xAI can have simple guardrails to avoid nudity, it can also implement guardrails to avoid all sorts of 'undressing' and placing people in suggestive poses.

The company's development team knows what to do and how to do it; they just need to DECIDE to do the right thing.

Everyone has been warned: if we cannot get AI companies to block "undressing" people (a very simple guardrail to implement), it sadly means that the state of AI governance is absolutely precarious.

Let's see how it goes...

I'll keep you posted.

-

👉 The best way to follow my work and my updates is through my newsletter. I invite you to join 88,600+ subscribers below.

The Italian Data Protection Authority

Co-founder of the AI, Tech & Privacy Academy (1,400+ participants), Author of Luiza’s Newsletter (88,000+ subscribers), Mother of 3

Tom Krumins

🚨 The State of New York is becoming the United States' strictest tech and AI regulator, and NOBODY is talking about it.

In December alone, NY enacted two important laws:

- The RAISE Act ("to require AI frameworks for AI frontier models, setting a nation-leading standard for AI transparency and safety")

- S4505/A5346 on social media warning labels ("to require social media platforms to display labels warning of the dangerous impact certain features can have on young users’ mental health")

It looks like NY wants to overtake California as the lead tech regulator in the U.S.

(Meanwhile, the federal government's goal seems to be to deregulate and block states that attempt to go against deregulation.)

Federalism is alive and thriving.

-

👉 To learn more about AI governance and AI's legal and ethical challenges, subscribe to my newsletter and join the 27th cohort of my training program (links below).

Most people haven't realized it, but AI-led reputational harm is on the rise.

Unlike search engines (where you can create an alert for your name), there is no way to control how AI will summarize or make up information about you.

That's what unregulated AI looks like...

Imagine a man in his 70s receiving an AI deepfake video of an attractive young girl asking him to be her boyfriend and to send money to help pay her bills.

Let's say he is lured and falls for the scam. The scammer continues. Maybe he will spend his life savings on this. One day, the scammer disappears, and the man wakes up to the sad reality of what happened.

His bank will deny reimbursement, and law enforcement will not be able to help.

Scammers are already getting rich by running the same cheap AI-powered scam over and over again, as they realize that the risk of being caught is very low.

AI platforms that help people create realistic deepfakes are freely available, under little to no oversight in most parts of the world.

The people most at risk are the elderly, and AI literacy efforts to educate them have been minimal.

This, ladies and gentlemen, is what the AI regulatory Wild West looks like.

For all those pushing for deregulation or insulting those who advocate for more rules and oversight: this could be your grandfather, your disabled aunt, or your old neighbor who can barely pay the bills.

AI harms are here, and we don't need to wait for 'AGI' or 'superintelligence' to do something.

AI literacy, AI governance, AI oversight, and AI regulation are needed now.

-

If you want to learn more about AI harms, the legal and ethical challenges of AI, and what pro-human policies, rules, and rights could look like, subscribe to my newsletter and read my articles (link below)

Connie Wang

🚨 BREAKING: The first legal settlement involving AI chatbot-related teen suicide is out.

This and the other ongoing lawsuits against CharacterAI and OpenAI over suicide, murder-suicide, and mental health harm are going to shape the field of AI liability.

So far, it is unclear to what extent AI companies are responsible for the harm (and have to compensate victims and their families), in cases involving emotional interactions with AI chatbots.

As I've written a few times in my newsletter, regardless of how people use AI systems, there should be built-in guardrails, warnings, and design elements to prevent harm.

Especially when children are involved and when it's a relatively new technology. AI literacy levels are still low and people are starting to learn to navigate the new risks. I'm adding below my recent essay on AI liability.

The terms of this settlement have not yet been disclosed.

I'll keep you posted.

-

👉 To receive my updates on AI's legal and ethical challenges, join my newsletter's 88,600+ subscribers (link below).

Co-founder of the AI, Tech & Privacy Academy (1,300+ participants), Author of Luiza’s Newsletter (88,000+ subscribers), Mother of 3

🚨 Most people haven't realized it yet, but the time to regulate AI is NOW.

These are still the early days of the ‘AI age,’ when AI becomes part of millions of people’s daily lives and pervasively influences them.

Whatever we decide to do now, however, will become systemic and potentially irreversible within 20 years.

If social media is any example, this is the time to set the tone, standards, values, rules, and rights that should be preserved and protected at any cost.

Now is the time to act rationally, collectively, and democratically to ensure that AI serves humans and humanity in the best possible way.

Read my article (and share if you agree) using the link below.

And why you should care | Edition #262 👋 Hi everyone, Luiza Jarovsky, PhD, here. Welcome to the 262nd edition of my newsletter, trusted by more than 

The Case for AI Regulation

Dismissing legitimate AI criticism as 'AI denialism' is inaccurate and ignores the nuances of the debate.

It's also an attempt to paint people offering valid counter-arguments to existing AI trends and policies as stupid.

Don't fall for this trap.

Denika M.

🚨 Anthropic is advancing legally questionable theories of AI personality to support an exceptionalist system for AI and WEAKER accountability frameworks for AI companies:

Three days ago, Anthropic launched Claude's new constitution. I did not even know Claude had an old constitution, so I checked it out.

While reading it, I was shocked at first.

Anthropic soaked the document in heavy doses of anthropomorphism and presented a pretentious, controversial, and legally inaccurate approach to the nature and social role of an AI model.

This did not seem to match the company's ambition to appear aligned with responsible AI principles and the mainstream AI governance discourse, as well as its interest in reigning over enterprise AI.

I was also surprised that this version of Claude's constitution received internal approval from the legal and PR teams for publication.

As I have written before, Anthropic has avoided social media drama and unproductive controversies. It has also seemed interested in legal compliance and ethical alignment. This has made them the favorite AI company of many lawyers and AI governance professionals I know.

Why would they risk their reputation publishing a document that will stir ‘AGI’ and ‘superintelligence’ rumors and provoke passionate negative reactions with sentences such as “Anthropic genuinely cares about Claude’s wellbeing”?

There is an AI race underway, and the people at Anthropic are very smart, so there must be a strategic reason to publish this document.

After thinking it through and watching the internet’s reaction to Claude’s new constitution (including some of the reactions to my teaser), I understood what is going on.

In a few words, under the guise of ‘full transparency,’ the company is advancing new, unpopular, and legally questionable theories of AI personality to support a parallel, weaker accountability framework for AI companies.

These theories of AI personality and the way Anthropic is framing them in Claude’s constitution fail to account for human rights, values, and societies, and they are being built into the company's AI model.

Strangely, their goal seems to be to create a higher-status, exceptionalist system for AI models and AI companies.

I do not think this is a good idea from an AI governance perspective, and I think it will backfire for Anthropic.

Let me explain to you how they are doing it: (...continues...)

👉 Continue reading my article and join 89,500+ subscribers using the link below.

Anthropic is advancing legally questionable theories of AI personality to support an exceptionalist system for AI and weaker accountability frameworks

Open article: Claude's Strange Constitution by Luiza Jarovsky, PhD • 3 min read

The entertainment industry is mostly anti-AI.

🚨 Most regulatory authorities in the world haven't realized yet, but general-purpose AI will NEVER be safe:

I have written several times in my newsletter that financial advice, psychotherapy, and storytelling for kids should never be offered as a single product or service.

This is intuitive for most people, and in addition to the social, cultural, and ethical reasons why it is true, there is also an important legal basis:

Financial advice, psychotherapy, and storytelling products or services will be used by different groups of people in different ways, present different risk profiles, and are governed by different sets of legal rules.

However, despite the existing consensus on the impracticality and potential unlawfulness of coupling financial advice, psychotherapy, storytelling, and other areas into a single product, general-purpose AI systems, starting with ChatGPT in November 2022, broke this rule.

Today, you can use general-purpose AI chatbots like ChatGPT, Gemini, Claude, and other competitors to obtain financial advice, psychotherapy, storytelling, and numerous other domains.

These various domains have little to nothing in common and would never be offered as a single physical product or service in an offline context.

Most regulatory authorities have not realized it yet (and hopefully this edition will offer more information and awareness about the topic), but this is a major obstacle to effective AI regulation and AI chatbot safety, and it must change.
When the same general-purpose AI chatbot is being used by:

- a man looking for intimate companionship,
- a teenage girl secretly asking about drugs and sex,
- a person with severe mental health issues looking for psychological support,

There will never be a single set of rules that will provide adequate service, protect the people involved, and address the context-specific risks.

When rules are inadequate for a product or service’s risk profile, people remain unprotected, and cases of harm happen more often (as we have observed over the past three years in AI, especially in the context of mental health harm and suicide).

Also, insufficient rules help AI companies avoid accountability for the harms their systems cause. They will argue that it is too difficult or "technically unfeasible" to prevent harm (another argument we have heard multiple times in the past three years).

The only solution to properly regulate AI chatbots - and I know many in AI will get angry with what I am saying, as every time I propose stricter AI rules, I am personally attacked - is to require AI companies to provide purpose-specific AI chatbots.

It is a big statement, but authorities and policymakers must realize that general-purpose AI chatbots will never be properly regulated, for the reasons I described above.

Authorities should require AI companies to define the target audience and (...continues...)

👉 Continue reading my article and join my newsletter's 89,200+ subscribers below.

Authorities should require AI companies to 1) limit the purposes of their AI systems; 2) respect sector-specific rules; 3) be held accountable for the

General Purpose AI Will Never Be Safe

🚨 Anthropic's new 'Constitution' for Claude is ethically and legally problematic.

Regulatory authorities and the AI community should be concerned.

I'll write more about it in my newsletter (subscribe below).

taplio

Luiza Jarovsky's Linkedin Analytics

Luiza Jarovsky

Performance ScoreLuiza Jarovsky's LinkedIn Performance

What is Luiza talking about?

Luiza Jarovsky's Best Posts (last 30 days)

Ready to grow your LinkedIn presence?

taplio

.css-6ytdur{background-image:linear-gradient(284deg, rgba(39, 43, 72, 1) 0%, rgba(33, 47, 183, 1) 100%);color:transparent;-webkit-background-clip:text;background-clip:text;}Luiza Jarovsky's Linkedin Analytics

Luiza Jarovsky

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Luiza Jarovsky's LinkedIn Performance

What is Luiza talking about?

Luiza Jarovsky's Best Posts .css-1emoo84{font-size:var(--chakra-fontSizes-md);font-weight:400;color:var(--chakra-colors-gray-600);}(last 30 days)

Ready to grow your LinkedIn presence?

Luiza Jarovsky's Linkedin Analytics

Performance ScoreLuiza Jarovsky's LinkedIn Performance

Luiza Jarovsky's Best Posts (last 30 days)