0. INTRODUCTION
In the news:
- DPA (Data Protection Authority) about “Clearview”: facial recognition company
- Australia wants to set age limit on social media
- Deepfakes: disinformation researchers have long warned that generative AI has the
ability to lower the bar for creating misleading content and threaten information around
elections.
- Online fraud (FSMA = watchdog)
- Legal sector: companies as Legalfly and Hechman
1. THE IMPORTANCE OF INTERDISCIPLINARITY – TECHNOLOGY + ECONOMY + SOCIETY
Importance of interdisciplinarity and legal impact of technology:
1. How does the technology work?
2. What is the business model? (economy)
3. Societal considerations
Technology is neither good nor bad, nor is it neutral.
Many different perspectives and of the roll of technology in society
Technology is … Autonomous Human-controlled
Neutral Determinism Instrumentalism
Value-laden Substantivism Critical theory
(technological) determinism: means that you believe that technology shapes society,
technology is neutral and that how individuals use technology does not matter. it has impact,
but its autonomous.
Instrumentalism: technology is neutral but you have an impact, as an individual, on how technology
is used (ex.: guns don’t kill people but people kill people people decide to use a tool in a specific
kind of way ; AI is not biased but society is)
1
,Substantivism: you do believe that values are inherently included in technology but as an
individual you don’t have an impact / the way individuals use it doesn’t matter (ex.: a gun is
made to kill people inherent something bad), Value-laden = involving subjective moral
evaluations
Critical theory: technology isn’t neutral, there is an inherent link with the values that exist in society
because it is people who create technology so the values of those people are automatically integrated
in technology + believe that it matters how people use technology, Human-controlled (people CAN
use technology in a way that wasn’t meant to be)
People look at technology in different ways vision is important for what regulation to use
Which considerations should policymakers, legislators, lawyers, … take into account?
= legal impact, regulation of technology:
Technical considerations
̶ How does the technology work?
̶ How is the hardware (physical parts: computer, smartphone, networks,…) and
software (apps, instructions to do certain things / code) designed?
̶ Which models are used?
‒ To show certain outputs to the user
̶ How are models trained?
̶ Which data is used? Personal data? ( GDPR)
̶ How is It obtained? (Clearview,…)
‒ Facial recognition doesn’t work good for all groups in the society (ex.
people of color)
̶ Who has designed the technology? (what is his vision?)
̶ How well does the technology work? (diagnoses, legal advice,…)
̶ How secure is the technology?
Economic considerations
̶ What is the business model (how are activities organised, how is value created, how is money made?)?
2
, ̶ Is a service free?
̶ Is a service paid by advertising?
̶ Do we pay with our data or input? “if something is free, you are the product”
̶ Who are the economic actors in the tech sector?
̶ ecosystems (YouTube, Android, Google Play, Gemini Alphabet); platforms
(two-sided; e.g. Uber / multi-sided; e.g. Uber Eats); companies; individuals?
̶ How much economic power do tech companies hold? (Google, Youtube,…)
̶ Which role do network externalities and lock-in play?
̶ more users = more value; market tipping (so many benefits that outweigh costs);
lock-in (a lot or users means a lot of data and switching (to another platform)
costs are too high)
Societal considerations
̶ Is the use of a certain technology desirable / acceptable / unacceptable? (AI,…)
̶ Ex.: emotion recognition systems
̶ Are there potential negative effects related to the use of the technology? for individuals? for society?
̶ How are (social, political, …) values shaped by technology?
̶ How do individuals use technology?
̶ How do cognitive abilities / biases influence the use of technology?
̶ privacy paradox: “people these days don’t care about privacy”, transparency
paradox, control paradox, the text to accept cookies is that long and boring that
people choose the benefit of going immediately on the service they want to use
̶ Do individuals need certain skills to use the technology?
̶ media literacy / digital literacy
Applying the considerations: ChatGPT:
- technology: these systems don’t think, quality of output is dependent on the unput (the
data): often the answer seems very convincing, although they can be totally wrong
- economics: some tools are free, others aren’t it’s not because it’s “free” they don’t
get paid for it (advertising, collecting data by you,…) you pay with your prompts (make
systems grow with what you insert in them)
- society: it could be very useful, but the answers are often biased is technology
biased/ is the data biased?
3
, AI doesn’t have access to all information / data ex. Belgian
case law
Can result in false answers that seem very
convincing
From societal point of view: some concerns: questions
about data, copyright related questions, questions about
biases, how can a chatbox become friends with people, very
easy to make fake news items and fake photos…
• Most AI applications come from one of the two leading countries in AI research and investment.
This means certain worldviews are favored in content processing and production.
• The entire executive team of OpenAI, the company behind ChatGPT, is under 40 years old. This
means that its language models reflect the specific ways of thinking and knowing of one
generation.
• Major chatbots have been trained on only about 100 of the world’s 7,000 natural languages, with
English as the primary source, given its predominance online. This means that 99% of the world's
languages are currently excluded from the giant virtual library that underlies the most popular
generative AI applications.
Clearly, these problems are linguistic, cultural, generational, and geopolitical.
Applying: Instagram:
Technology: what you see is decided by an (often unclear)
algorithm. You often see more (untrue) sensational content that
can lead you in a negative spiral/specific type of content in a
short space of time
Digital Services Act: platforms need to create a feed that is
NOT personalized (ex. chronological)
Economics: advertising based so it tries to keep you as long
as possible
Economics: Instagram is part of the Meta ecosystem: data can be easily shared to other
platforms
Society: Lots of people experience lock-in effect: even though you don’t agree with how the
company acts, you stay because all your friends have it / you don’t want to miss out on things…
Society: negative effects for protection of personal data & mental health, impact of spreading of
misinformation
4
In the news:
- DPA (Data Protection Authority) about “Clearview”: facial recognition company
- Australia wants to set age limit on social media
- Deepfakes: disinformation researchers have long warned that generative AI has the
ability to lower the bar for creating misleading content and threaten information around
elections.
- Online fraud (FSMA = watchdog)
- Legal sector: companies as Legalfly and Hechman
1. THE IMPORTANCE OF INTERDISCIPLINARITY – TECHNOLOGY + ECONOMY + SOCIETY
Importance of interdisciplinarity and legal impact of technology:
1. How does the technology work?
2. What is the business model? (economy)
3. Societal considerations
Technology is neither good nor bad, nor is it neutral.
Many different perspectives and of the roll of technology in society
Technology is … Autonomous Human-controlled
Neutral Determinism Instrumentalism
Value-laden Substantivism Critical theory
(technological) determinism: means that you believe that technology shapes society,
technology is neutral and that how individuals use technology does not matter. it has impact,
but its autonomous.
Instrumentalism: technology is neutral but you have an impact, as an individual, on how technology
is used (ex.: guns don’t kill people but people kill people people decide to use a tool in a specific
kind of way ; AI is not biased but society is)
1
,Substantivism: you do believe that values are inherently included in technology but as an
individual you don’t have an impact / the way individuals use it doesn’t matter (ex.: a gun is
made to kill people inherent something bad), Value-laden = involving subjective moral
evaluations
Critical theory: technology isn’t neutral, there is an inherent link with the values that exist in society
because it is people who create technology so the values of those people are automatically integrated
in technology + believe that it matters how people use technology, Human-controlled (people CAN
use technology in a way that wasn’t meant to be)
People look at technology in different ways vision is important for what regulation to use
Which considerations should policymakers, legislators, lawyers, … take into account?
= legal impact, regulation of technology:
Technical considerations
̶ How does the technology work?
̶ How is the hardware (physical parts: computer, smartphone, networks,…) and
software (apps, instructions to do certain things / code) designed?
̶ Which models are used?
‒ To show certain outputs to the user
̶ How are models trained?
̶ Which data is used? Personal data? ( GDPR)
̶ How is It obtained? (Clearview,…)
‒ Facial recognition doesn’t work good for all groups in the society (ex.
people of color)
̶ Who has designed the technology? (what is his vision?)
̶ How well does the technology work? (diagnoses, legal advice,…)
̶ How secure is the technology?
Economic considerations
̶ What is the business model (how are activities organised, how is value created, how is money made?)?
2
, ̶ Is a service free?
̶ Is a service paid by advertising?
̶ Do we pay with our data or input? “if something is free, you are the product”
̶ Who are the economic actors in the tech sector?
̶ ecosystems (YouTube, Android, Google Play, Gemini Alphabet); platforms
(two-sided; e.g. Uber / multi-sided; e.g. Uber Eats); companies; individuals?
̶ How much economic power do tech companies hold? (Google, Youtube,…)
̶ Which role do network externalities and lock-in play?
̶ more users = more value; market tipping (so many benefits that outweigh costs);
lock-in (a lot or users means a lot of data and switching (to another platform)
costs are too high)
Societal considerations
̶ Is the use of a certain technology desirable / acceptable / unacceptable? (AI,…)
̶ Ex.: emotion recognition systems
̶ Are there potential negative effects related to the use of the technology? for individuals? for society?
̶ How are (social, political, …) values shaped by technology?
̶ How do individuals use technology?
̶ How do cognitive abilities / biases influence the use of technology?
̶ privacy paradox: “people these days don’t care about privacy”, transparency
paradox, control paradox, the text to accept cookies is that long and boring that
people choose the benefit of going immediately on the service they want to use
̶ Do individuals need certain skills to use the technology?
̶ media literacy / digital literacy
Applying the considerations: ChatGPT:
- technology: these systems don’t think, quality of output is dependent on the unput (the
data): often the answer seems very convincing, although they can be totally wrong
- economics: some tools are free, others aren’t it’s not because it’s “free” they don’t
get paid for it (advertising, collecting data by you,…) you pay with your prompts (make
systems grow with what you insert in them)
- society: it could be very useful, but the answers are often biased is technology
biased/ is the data biased?
3
, AI doesn’t have access to all information / data ex. Belgian
case law
Can result in false answers that seem very
convincing
From societal point of view: some concerns: questions
about data, copyright related questions, questions about
biases, how can a chatbox become friends with people, very
easy to make fake news items and fake photos…
• Most AI applications come from one of the two leading countries in AI research and investment.
This means certain worldviews are favored in content processing and production.
• The entire executive team of OpenAI, the company behind ChatGPT, is under 40 years old. This
means that its language models reflect the specific ways of thinking and knowing of one
generation.
• Major chatbots have been trained on only about 100 of the world’s 7,000 natural languages, with
English as the primary source, given its predominance online. This means that 99% of the world's
languages are currently excluded from the giant virtual library that underlies the most popular
generative AI applications.
Clearly, these problems are linguistic, cultural, generational, and geopolitical.
Applying: Instagram:
Technology: what you see is decided by an (often unclear)
algorithm. You often see more (untrue) sensational content that
can lead you in a negative spiral/specific type of content in a
short space of time
Digital Services Act: platforms need to create a feed that is
NOT personalized (ex. chronological)
Economics: advertising based so it tries to keep you as long
as possible
Economics: Instagram is part of the Meta ecosystem: data can be easily shared to other
platforms
Society: Lots of people experience lock-in effect: even though you don’t agree with how the
company acts, you stay because all your friends have it / you don’t want to miss out on things…
Society: negative effects for protection of personal data & mental health, impact of spreading of
misinformation
4
