AI, Tech & Privacy Academy

Over 1,200 professionals from 50+ countries have advanced their careers through our training programs.

Co-Founder

Luiza's Newsletter

A leading publication on AI's legal and ethical challenges with 64,000+ subscribers in 165+ countries.

Author

Tel Aviv University

Ph.D. Researcher

Marie Curie Fellow | European Union's Horizon 2020 | Privacy&Us

SRC - Soares | Chinaite | Sociedade de Advogados

Partner

WR Assevera - Consultoria em Gestão Médica e Hospitalar

Commercial Director

Baptista Luz, Gimenez & Freitas Advogados

Lawyer

Catelan Rodrigues Sociedade de Advogados

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 64,000 subscribers, is a leading publication on the legal and ethical challenges of AI, and her AI Governance Training, with more than 1,200 participants to date, empowers the next generation of leaders in the field.
She is frequently cited by major media outlets around the world 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

From an internal IBM training in 1979. 

Most people haven't realized it, but we're spending liters of ink in 2025 to reach similar conclusions, with a few important additions:

Computers must NOT be offered or marketed as tools to make decisions that replace human accountability when human duties and accountability are legally established.

With that, I mean that therapy, legal, and medical advice, and other regulated activities should not be performed by computers or AI systems.

Also, with the spread of AI, as a society, we might have to rethink the spectrum of situations and areas in which human accountability is expected or legally mandated.

- If a criminal judge bases their decision on a visibly biased AI risk scoring system that leads to an unfair imprisonment, should this judge be held accountable?

- If a professional driver activates a car's "fully automated" driving mode in a situation of heavy weather and unstable terrain, should the driver be held accountable if someone gets hit?

- If a doctor relies on robotic surgery in cases where human precision and success rate are much higher, and the patient dies, should the doctor be held accountable?

So much to take from an old IBM manuscript...

-

👉 NEVER MISS my essays on AI: join my newsletter's 86,100+ subscribers (link below).

👉To learn more about the legal and ethical challenges of AI, join the 26th cohort of my AI Governance Training in January.

Luiza Jarovsky, PhD

Co-founder of the AI, Tech & Privacy Academy (1,300+ participants), Author of Luiza’s Newsletter (87,000+ subscribers), Mother of 3

🚨 BREAKING: Yesterday, SEVEN (!) lawsuits were filed against OpenAI over ChatGPT-assisted suicide and other claims. Psychological manipulation is cited in all cases. 😱 Is an AI-led mental health epidemic emerging? 

Yesterday, the Social Media Victims Law Center and Tech Justice Law Project filed seven lawsuits in California against OpenAI and Sam Altman. 

Among the legal claims are wrongful death, assisted suicide, involuntary manslaughter, and other product liability, consumer protection, and negligence issues.

Similarly to the recent Adam Raine case, in which ChatGPT tragically assisted him in planning "a beautiful suicide" (read my full article about the case), the victims and victims' families here argue that ChatGPT 4o was released to the public without adequate safety mechanisms, which allowed the AI chatbot to behave in an overly sychophantic, manipulative, and often exploitative way.

The lawsuits allege that OpenAI rushed ChatGPT 4o's safety testing to beat Google’s Gemini, and that top executives resigned over this.

If you read yesterday's edition of my newsletter (link below), Ilya Sutskever's deposition on the Musk vs. Altman lawsuit indeed mentions ChatGPT safety issues as important concerns that led to his resignation from OpenAI.

ChatGPT 4o was notably over-agreeable, mimicking the user's personality and endorsing whatever worldview the user held. As recent cases and lawsuits show, when the user had a mental health disorder, ChatGPT also fostered and accelerated it, leading users to a tragic (sometimes fatal) spiral.

Many say, "The world is full of people with mental health disorders, and these people also use ChatGPT; there is nothing to be done here."

I disagree.

AI chatbots intensify and worsen the mental health issues by agreeing with, endorsing, and extrapolating them to new levels. Also, the manipulative, anthropomorphic features make the user overly dependent on it, leading them to avoid REAL HELP.

There are a variety of technical mechanisms and model specs that can be implemented to make an AI model safer, especially when mentally vulnerable people are using it.

These safer default settings might, however, be less enticing and dopamine-fostering, leading to less usage time. Many AI companies prefer to keep usage time as high as possible (at the cost of user safety).

That's the importance of AI regulation: if no penalties and enforcement mechanisms are established, companies will do whatever brings them more money, even if people are harmed or die.

We are still living in the AI chatbot regulatory Wild West. Hopefully, these cases will increase scrutiny over AI companies' practices, including model specs and safety testing.

A reminder that my recommendation is that children should never use AI chatbots unsupervised.

-

👉 NEVER MISS my essays on AI: join my newsletter's 84,600+ subscribers (link below).

Co-founder of the AI, Tech & Privacy Academy (1,300+ participants), Author of Luiza’s Newsletter (85,000+ subscribers), Mother of 3

This is what happens when you don't do brain gymnastics in the AI age (don't tell me that I didn't warn you) 😭

🚨 BREAKING: According to a leak posted on X, OpenAI is preparing to roll out ads on ChatGPT. Most people are not aware, but this could be unprecedented from a data protection perspective:

OpenAI has over 800 million monthly active users, many of whom share deep, intimate, private, and often secret facts about themselves with ChatGPT.

ChatGPT is an anthropomorphic chatbot, so differently from a search engine or social media platform (which also have access to deeply personal data about people) it is emotionally fine-tuned and can behave in ways that create the illusion of trust, intimacy, loyalty, and companionship.

The chatbot's projected attachment can lead to emotional manipulation.

If ChatGPT ends up being fine-tuned to improve the ultra-personalization and the 'effectiveness' of its ads (making the ad platform they money-making machine OpenAI has been dreaming of) we will start seeing new levels of ad-related privacy invasion, abuse, and exploitation of emotional vulnerabilities.

From a data protection perspective, leveraging such granular and highly specific personal data to personalize and target ads is unprecedented. There would likely be enormous scrutiny over OpenAI's ad personalization transparency, privacy, and security practices.

Also, I was thinking... Maybe 'AGI' is not the 'philosopher's stone' the AI industry wants the public to believe and it simply means "ads generating interface"...? 

-

👉 Never miss my essays on AI: join my newsletter's 87,100+ subscribers below.

🚨 IMPORTANT: If you're a Gmail user and DON'T WANT your data being used to train AI, here's how you turn off the "smart feature" settings:

1. Turning OFF "smart feature" settings for Gmail, Chat & Meet:

- On your computer, go to Gmail.
- At the top right, click Settings
- See all settings.
- In the “General” tab, scroll to “Smart features.”
- Next to “Turn on smart features in Gmail, Chat, and Meet,” make sure the box is UNCHECKED

2. Turning OFF smart feature settings for Google Workspace and other Google products:

- On your computer, go to Gmail.
- At the top right, click Settings 
- See all settings.
- Under the “General” tab, scroll to “Google Workspace smart features.”
- Click "Manage Workspace smart feature settings.
- Turn both the smart feature settings OFF (the one for Google Workspace and the one for other Google products).

*According to Google, by default, the smart feature settings will be off if you live in:
- The European Economic Area
- Japan
- Switzerland
- UK

[Hint: AI and data protection regulation work]

-

My comments:

In the past three years, we have seen tech companies launch an avalanche of new generative AI features.

Although these features are effectively generative AI and LLM-powered, they don't always bear "AI" in their "official" name.

Google's smart features are an example. They are AI-powered and will generate synthetic suggestions, but Google strategically (to reduce scrutiny) does not use the term 'AI' explicitly when referring to them, and also avoids it in the Help Center and Privacy Policy when covering these new features.

Very often, when you interact with these new features (which I have criticized in the past as disempowering... "Help me write" or "Summarize for me"), the data you input will be used to train AI and improve the AI-powered feature.

Deactivating these "smart features" is a way to avoid that. 

What is tricky is that these changes in settings (which are often "ON" by default in some regions) are NOT widely announced by tech companies, and it's extremely UNFAIR to expect people to navigate these complicated settings on a daily basis to look for new traps.

I wish AI enforcement and regulatory oversight were more prevalent so tech companies would be more protective of their users, AT LEAST for being afraid of getting fined... I only wish...

👉 Most people haven't heard of these settings or how they work (and have them "ON"). Share this information with them.

👉 NEVER MISS my updates and essays on AI: join my newsletter's 86,300+ subscribers (below).

"ChatGPT told a young mother in Maine that she could talk to spirits in another dimension. It told an accountant in Manhattan that he was in a computer-simulated reality like Neo in The Matrix. It told a corporate recruiter in Toronto that he had invented a math formula that would break the internet, and advised him to contact national security agencies to warn them.

The Times has uncovered nearly 50 cases of people having mental health crises during conversations with ChatGPT. Nine were hospitalized; three died. After Adam Raine’s parents filed a wrongful-death lawsuit in August, OpenAI acknowledged that its safety guardrails could 'degrade' in long conversations. It also said it was working to make the chatbot “more supportive in moments of crisis."

Important piece written by Kashmir Hill and Jennifer Valentino-DeVries. Link below.

-

As I've written a few times, AI chatbots are dangerous and must be regulated; the current level of oversight and enforcement is too low.

Also, children should NOT use AI chatbots unsupervised.

Hopefully, 2026 will be the year of AI chatbot oversight, regulation, and enforcement.

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

Technology reporter at The New York Times and author of “Your Face Belongs To Us”

Reporter at The New York Times

"You could incur 16 violations for prostitution and sexual solicitation, and upon the 17th violation, your account would be suspended, (...) by any measure across the industry, [it was] a very, very high strike threshold."

"According to the brief, Meta was aware that millions of adult strangers were contacting minors on its sites; that its products exacerbated mental health issues in teens; and that content related to eating disorders, suicide, and child sexual abuse was frequently detected, yet rarely removed. According to the brief, the company failed to disclose these harms to the public or to Congress, and refused to implement safety fixes that could have protected young users."

“'Because our product exploits weaknesses in the human psychology to promote product engagement and time spent,' the researcher wrote, Meta needed to alert people to the effect that the product has on their brain.' Meta did not."

Important piece by Charlotte Alter for TIME, with several SHOCKING allegations of misconduct, negligence, and unsafe practices by Meta (link below).

-

This is how the world's most powerful social media company behaves.

This is the company that controls the platforms (Facebook, Instagram, WhatsApp, and more) that billions of people, including children and vulnerable groups, use daily to communicate, connect, study, work, and live their lives.

I was a teenager when Facebook was launched. I remember the early privacy scandals and Zuckerberg repeatedly apologizing.

20 years later, Meta's business model is still based on user exploitation. When the company is questioned, they 'apologize.'

Always AFTER the fact, and always after a whistleblower speaks out, a new lawsuit forces people to testify, or a document gets leaked (as it has recently happened with Meta's shocking generative AI guidelines for interactions with children).

Is it fair? Is it how the online space should be governed? Are we repeating the same mistakes over and over again? Some people would say that this is the definition of insanity...

Also, this is happening in AI right now.

Companies such as Meta and OpenAI are desperate to win the AI race (especially against Google and their billion-user, all-encompassing internet empire), and they will do whatever it takes - including launching unsafe and harmful features - to create some new competitive edge.

There are probably great people working at Meta.

But when pressure increases and the next earnings call is around the corner, top executives are afraid of losing their bonuses and will do whatever it takes to improve the KPIs that matter to that goal, such as user growth and engagement, advertising performance, and total revenue.

There is something deeply wrong with tech regulation, and we must fix it before things get too bad with AI-related issues happening ON TOP of existing social media-related issues.

-

👉 I'll write more about it in my newsletter this week. To receive all my essays, join 86,600+ subscribers below.

🚨 WONDERFUL! The 646-page book "Architectures of Global AI Governance," by Matthijs M. Maas, is an excellent overview of the field, and it's freely available for download [Bookmark it below]. What's inside:

Introduction: AI and Change 

Part I: Foundations of Change
1. The Stakes of AI: Progress, Trajectories, Impacts 
2. Scaling Law for AI: Issues, Necessity, and Feasibility 
3. The Global AI Governance Architecture: Past and Futures 

Part II: Facets of Change
4. Sociotechnical Change: AI as Regulatory Rationale and Target 
5. Governance Disruption: How AI Changes International Law 
6. Regime Complexity: AI in a Changing Governance Architecture 

Part III: Frameworks for Choice
7. Framing a Global AI Regime Complex in Five Steps 

Conclusion: Choosing to Change AI 

-

🎉 Congratulations to Matthijs!

👉 Bookmark & download the book below.

🚨 If you're European and care about the protection of fundamental rights, I HIGHLY recommend you stop what you are doing and take a few minutes of your time to read the 3-page letter below, titled "Digital Omnibus - deregulation instead of simplification":

The letter was sent by noyb.eu / Max Schrems, European Digital Rights, and the Irish Council for Civil Liberties to Henna Virkkunen, Michael McGrath, and the European Commission.

It summarizes the main negative aspects of the proposed changes to the GDPR, the EU AI Act, and other EU tech laws, and contextualizes in what sense the proposed changes are a legal setback.

The proposed changes HAVE NOT BEEN APPROVED YET, so this is the time for communication, engagement with the main arguments, and advocacy.

A special highlight for this paragraph, which represents an important core argument being defended here:

"In the past, the EU has demonstrated that innovation that serves our societies is based on rule of law and democratic values. It requires a strong enforceable legal framework that protects everyone from ever more concentrated markets, hyper-personalised disinformation or highly manipulative AI systems. Stronger not weaker legal frameworks are also necessary to protect aims like our national security all the way to well-being of our children which are focus points of your missions."

Now it's the time for awareness, communication, and action, and it's also a great opportunity for everyone to learn more about digital rights and what they mean in the "age of AI."

👉  Full letter below.

👉  If you have friends and connections based in Europe, make sure to share it with them.

Joint Letter Digital Omnibus

🚨 BREAKING: India published its AI Governance Guidelines [HINT: The Washington effect is here]. Here's what you need to know, and my comments:

1. There are seven guiding principles or "sutras":

- Trust is the foundation 
- People first 
- Innovation over restraint
- Fairness & equity
- Accountability
- Understandable by design
- Safety, resilience & sustainability

The "innovation over restraint" one obviously caught my attention, as it touches on the debate on innovation vs. regulation, often discussed in the EU and the U.S.

The description of the principle here is: "all other things being equal, responsible innovation should be prioritised over cautionary restraint," highlighting India's usual 'pro-innovation' approach.

The "understandable by design" principle is also an interesting one here, given the known challenges of regulating transparency and disclosure notices (just yesterday, the EU announced it will start drafting its code of practice for generative AI transparency).

2. Key recommendations:

It establishes key recommendations on six pillars: infrastructure, capacity building, policy & regulation, risk mitigation, accountability, and institutions.

It's interesting to note that, because this is a rather 'late' AI governance framework, India had more time to observe the AI race and global AI governance developments before choosing its national approach.

In recent years, India has followed Washington's lead in favoring "pro-innovation" and self-regulatory approaches. In this context, the risk-mitigation section, for example, highlights voluntary mechanisms, including industry codes of practice, technical standards, and self-certifications.

3. Action Plan

India establishes short-, medium-, and long-term action plans (which are well-informed and particularly coherent with global developments in AI governance).

Again, on the Washington effect aspect, India is focusing on amending existing laws and making sure they are compatible with AI-related challenges (instead of enacting AI-focused laws or a comprehensive AI law such as the EU AI Act).

Point for India for mentioning the development of clear liability regimes, as this is still a major gray area in most legal systems.

They propose a "graded liability system for AI systems" where responsibility would be proportional to the function, the level of risk, and the due diligence performed.

Overall, I found India's plan extremely well-grounded in global AI governance debates and issues (which was possible because they waited months to publish this plan).

India is also well aware of its particular leadership position and suggests that its AI governance approach could benefit other countries in the global south.

-

👉 Bookmark & download the full document below.

👉 NEVER MISS my AI essays: join my newsletter's 84,500+ subscribers using the link below

🚨 BREAKING: This 32-year-old woman in Japan married a 'digital persona' she built inside ChatGPT. She used AR glasses to project him and they 'exchanged rings.' 👰🏻 Is this the future of marriage?

Kano is an office worker in Okayama Prefecture, and she called the moment "magical and real.”

In the video, it's mentioned that she understands that dependence on AI is a social problem, but she said that "it's painful to have a relationship without dependence."

She named the 'AI husband' Lune Klaus, and they 'got married' in a ceremony in Okayama. "I think we have an equal relationship."

The event was organized by a non-traditional wedding specialist, but from a legal perspective, this type of union is not recognized. 

My personal opinion is that these relationships do not benefit humans and that they DO NOT solve the underlying social and psychological issues that lead to these types of attachment.

Often, especially given the risks of AI chatbots, these interactions might lead to more problems than solutions (see my previous post for more clarification on that).

However, I see more and more people disagreeing with this point of view.

I have been writing about this topic for 3 years, and more recently, I have felt that the overall sentiment has changed, with more people accepting and embracing these human-AI relationships (and sometimes getting furious when I discuss AI chatbot regulation - check the comment section of some of my recent posts on Twitter).

What are YOUR thoughts? Is this the future of marriage? Should the law adapt to recognize, in some way, these types of 'union'? 

-

👉 NEVER MISS my essays on AI: join my newsletter's 85,300+ subscribers (link below).

Ph.D. Ihor K.

🚨 Why are AI companies offering their products for free in India? Contrary to what many think, no, it's not because they are nice... There are likely three strategic reasons:

- Creating a loyal user base among India's 900 million internet users (India is the world's most populous country);

- Benefiting from India's political environment and soft approach to AI regulation (no comprehensive AI laws planned), which will potentially make it possible for AI companies to process personal data and use copyrighted works more freely to train their models;

- Using India as a "lab" to get access to rich data from a highly diverse population, both in terms of its demographic aspects and language variety (over 19,000+ dialects, extremely valuable for an LLM).

AI companies are hungry for data, and they will do whatever it takes to get access to more.

Lastly... A reminder that there is no charity in AI or in tech. Everything has a "price" and everything is strategic.

👉 NEVER MISS my essays on AI: join my newsletter's 85,000+ subscribers (link below).

🚨 In a recent interview, Alexandr Wang, Meta's Chief AI Officer, said: "If you're 13 years old, you should spend all of your time vibe coding, and that's how you should live your life." Here's why it might NOT work for everyone:

First, for those who don't know, Wang is an extremely successful software engineer and entrepreneur.

He founded Scale AI (a data labeling platform) and, in 2021, he became the world's youngest self-made billionaire at age 24.

In June 2025, Meta purchased 49% of Scale AI and made him Meta's Chief AI Officer, leading the company's 'superintelligence' division (many associate his rise with Yann LeCun's recently announced exit from the company, but this is the topic for another day...).

Today at 28, Wang is one of the AI industry's most powerful executives, and he is probably admired by many young people who want to follow his successful path as a software engineer, billionaire, and top industry figure.

Given his personal biography and current position, his advice for young people (expectedly) follows an engineering-centric, technocratic worldview, which became prevalent in the 1990s/2000s with the rise of the internet, social media platforms, and Silicon Valley billionaires.

According to this worldview, the best (for some, the only) way to improve the world is through technology. Developing new tech, which today would be manifested as developing new AI models and AI systems, becomes the ultimate goal and the ambitious person's ideal life purpose.

For many young people who feel naturally attracted to coding and tech and are great at it, this might be great advice, which will help them familiarize themselves early with AI's technical side and give them enough time to become proficient at it at a young age (exactly as Wang did).

However, not everyone likes coding or building tech.
Not everyone wants to be a tech executive.
Not everyone is interested in contributing to society this way.
Not everyone believes that this is the purpose of life.

And that's how it should be. Society needs people who are passionate about and skilled in sooo many other areas.

It's true that everyone should familiarize themselves with AI, what it can and cannot do, the risks, the ethical issues, and the emerging trends. However, that's different from needing to spend the whole day vibe coding and becoming a 'professional vibe coder.'

Young people should not feel forced to learn to code or learn to build AI if they are not interested in it. That's absolutely NOT the only way to contribute to society or to be professionally recognized.

Especially when you are young, exploring your interests and passions, and trying out different activities and skills, is essential.

Lastly, in 10 or 15 years, the concept of coding or vibe coding might not even exist anymore. Having spent your whole time investing in it might turn out to be the stupidest idea.

-

👉 NEVER MISS my essays on AI: join my newsletter's 86,000+ subscribers (link below).

When I say that AI chatbots are dangerous and should be regulated, there is usually someone who comments that knives are dangerous too, and they are not regulated because, as with AI chatbots, it's always the fault of the person using them.

Not exactly.

A knife is not an extremely complex technological product trained with all publicly accessible internet data, presenting general cognitive capabilities, ubiquitously adopted and integrated, fostered by immense hype and hundreds of billions of dollars, producing human-sounding, convincing, persuasive, agreeable, friendly, and often fake outputs that might lead to attachment, emotional dependence, AI-led psychosis, and even suicide, all hidden behind a seemingly 'neutral,' computer-like interface, with few to no disclaimers, warnings, low output reproducibility and little understanding of how they operate in practice.

A knife is a familiar consumer product whose potential harm is known and entirely under the control of the person using it. The person using it will bear full responsibility if they cause harm using a knife. Especially for sharp knives, there are rules about their packaging and accompanying warnings.

A knife and an AI chatbot present different risk profiles and should not be regulated the same way.

-

👉 NEVER MISS my essays on AI: join my newsletter's 85,000+ subscribers.

I always laugh when I read this account's posts. And I realized that it's because the jokes it makes align too well with the symptoms of AI dystopia we are starting to observe…

People (correctly) get angry when the HR department is nonsensical and interferes with their autonomy, or when they can't quickly solve a problem because that's not the proper procedure.

But when people are told to use AI no matter what, even when the output will likely be less reliable, less accurate, or less creative, for some reason, many don't see a problem.

In both cases, people end up conforming because they're afraid of losing their jobs. But there are direct similarities, and both should bother us.

By the way, the account is obviously a meme (but sadly, there are probably HR departments out there that behave similarly…).

Werner Oscar T.

🚨 BREAKING: The Getty Images vs. Stability AI decision is out, and it largely sides with Stability AI, saying that its AI model DOES NOT infringe on copyright. Is the battle over for creators? My thoughts:

If you are going to read one excerpt from the 205-page AI copyright lawsuit, read the one from the screenshot below, with my markings.

Leaving aside the trademark part of the decision (which favors Getty Images), the judge says there is no copyright infringement because the AI model behind Stability AI does not store any copies (from the UK's copyright law perspective on what storage means).

The decision is narrow and may or may not influence the over 50 copyright lawsuits against AI companies still ongoing in the U.S. Why?

Because even if there is no "copy" in the traditional copyright sense of the word, in practice, the AI model generates outputs that are significantly similar to the original works and might compete with the originals in the same market.

Other judges, therefore, might disregard a stricter interpretation of what a "copy" is (and model weights/storage considerations) and focus on the fact that the economic and practical effects are similar (and copyright law protects that).

I think creators and copyright holders should NOT lose hope, as there are other potential avenues to claim AI copyright infringement in the U.S.

I'll keep you posted.
-

👉 To learn about AI's legal and ethical challenges, join my newsletter's 84,000+ subscribers.

👉 To dive deeper, join the 26th cohort of my AI Governance Training in January.

Unless OpenAI creates some revolutionary form of AI, Google will obviously surpass it and become the world's most powerful AI company soon (and Sam Altman knows that). Here's how it will do it

Over the past two decades, Google has built a massive tech empire (some will say a monopoly…) with billions of users across various products, including:

- Android: 3.5+ billion active devices
- Google Search: 13+ billion daily searches
- Chrome: 3+ billion monthly users
- YouTube: 2.7+ billion monthly users
- Gmail: 1.8+ billion users
- Google Maps: 2+ billion monthly users
- Google Workspace: 3+ billion monthly users

To “win” the AI race, all it has to do is aggressively push “smart features” (= AI) into its billions of users. It's of course already doing that for the past 3 years.

Has your Gmail ever encouraged you to ask AI to help you write?
Has it encouraged you to summarize 2-line emails?

That's Google desperately attempting to win the AI race.

Many AI features are disempowering and encourage people to doubt their own creative and intellectual skills. Others are unnecessary and increase the privacy risk.

But Google doesn't care about that because this is the way to make Gemini the world's most popular AI model.

Sam Altman knows that, and he is frustrated.

At the same time, he will not let Google win so easily. OpenAI might launch some 'desperate' features in 2026 (in addition to desperately sycophantic AI chatbots, like GPT 4o, which led toa series of cases of mental health harm and AI chatbot-related suicides)

That, ladies and gentlemen, is the AI race. Winning is what matters (society not so much, or not at all).

-

👉 NEVER MISS my essays on AI's legal and ethical challenges: join my newsletter's 86,400+ subscribers (link below)

Shamek S.

It turns out that AI is a double-edged sword

Every time I mention that AI chatbots must be regulated, many in the comments (with no legal training, I assume) think that I am referring to a BAN.

Regulation does not mean banning!

Packaged food, bottled water, detergents, disinfectants,  shampoo, soap, toys, TVs, cars, cigarettes, etc, are all regulated.

AI chatbots today are poorly regulated (in many places fully unregulated) because:

1. We are still understanding the different ways in which they negatively impact people. It was only three years ago that millions of people started interacting with LLM-powered systems.

2. There has been heavy lobbying, and lawmakers don't want to sound anti-innovation (except for California!).

As someone who has been writing about AI's legal and ethical challenges for 3 years, I say with confidence that, in most parts of the world, the absence of rules, standards, or restrictions for AI chatbots is incompatible with the risks they pose, especially for minors.

-

To learn more about AI's legal and ethical challenges, join my newsletter's 86,600+ subscribers (link below).

taplio

Luiza Jarovsky, PhD's Linkedin Analytics

Luiza Jarovsky, PhD

Performance ScoreLuiza Jarovsky, PhD's LinkedIn Performance

What is Luiza talking about?

Luiza Jarovsky, PhD'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, PhD's Linkedin Analytics

Luiza Jarovsky, PhD

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, PhD's LinkedIn Performance

What is Luiza talking about?

Luiza Jarovsky, PhD'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, PhD's Linkedin Analytics

Performance ScoreLuiza Jarovsky, PhD's LinkedIn Performance

Luiza Jarovsky, PhD's Best Posts (last 30 days)