South Korea Unveils Comprehensive AI Laws Amid Rising Global Concerns, But Critics Say They Fall Short.
The country has taken a bold step into regulating artificial intelligence (AI) by launching what is billed as the most comprehensive set of laws anywhere in the world. The move aims to position South Korea as one of the leading AI powers globally, but it's facing significant pushback from both tech startups and civil society groups.
The new legislation, dubbed the "AI Basic Act," requires companies providing AI services to label their creations with invisible digital watermarks for clearly artificial outputs such as cartoons or artwork. For realistic deepfakes, visible labels are required. The law also mandates risk assessments and documentation for high-impact AI systems used in medical diagnosis, hiring, and loan approvals.
However, the threshold for extremely powerful AI models is set so high that government officials acknowledge no models worldwide currently meet it. Companies that violate the rules face fines of up to 30 million won (£15,000), but a grace period of at least a year before penalties are imposed.
Critics argue that the law does not go far enough in protecting citizens from AI risks. A survey found that 98% of AI startups were unprepared for compliance, and experts warn of competitive imbalance. All Korean companies face regulation regardless of size, while only foreign firms meeting certain thresholds must comply.
The push for regulation comes against a uniquely charged domestic backdrop. South Korea accounts for 53% of all global deepfake pornography victims, according to a recent report. The law's origins predate this crisis, but provisions were repeatedly stalled due to industry interests over citizen protection concerns.
Civil society groups maintain that the new legislation provides limited protection for people harmed by AI systems. Four organisations issued a joint statement arguing that the law contains almost no provisions to protect citizens from AI risks. The groups noted that while the law stipulated protection for "users," those users were hospitals, financial companies, and public institutions that use AI systems, not people affected by AI.
The country's human rights commission has criticised the enforcement decree for lacking clear definitions of high-impact AI, noting that those most likely to suffer rights violations remain in regulatory blind spots. Experts say South Korea chose a different path from other jurisdictions, opting for a more flexible, principles-based framework that prioritises "trust-based promotion and regulation."
The effectiveness of this approach remains to be seen, but the global community is watching closely as countries grapple with rapidly advancing technologies.
The country has taken a bold step into regulating artificial intelligence (AI) by launching what is billed as the most comprehensive set of laws anywhere in the world. The move aims to position South Korea as one of the leading AI powers globally, but it's facing significant pushback from both tech startups and civil society groups.
The new legislation, dubbed the "AI Basic Act," requires companies providing AI services to label their creations with invisible digital watermarks for clearly artificial outputs such as cartoons or artwork. For realistic deepfakes, visible labels are required. The law also mandates risk assessments and documentation for high-impact AI systems used in medical diagnosis, hiring, and loan approvals.
However, the threshold for extremely powerful AI models is set so high that government officials acknowledge no models worldwide currently meet it. Companies that violate the rules face fines of up to 30 million won (£15,000), but a grace period of at least a year before penalties are imposed.
Critics argue that the law does not go far enough in protecting citizens from AI risks. A survey found that 98% of AI startups were unprepared for compliance, and experts warn of competitive imbalance. All Korean companies face regulation regardless of size, while only foreign firms meeting certain thresholds must comply.
The push for regulation comes against a uniquely charged domestic backdrop. South Korea accounts for 53% of all global deepfake pornography victims, according to a recent report. The law's origins predate this crisis, but provisions were repeatedly stalled due to industry interests over citizen protection concerns.
Civil society groups maintain that the new legislation provides limited protection for people harmed by AI systems. Four organisations issued a joint statement arguing that the law contains almost no provisions to protect citizens from AI risks. The groups noted that while the law stipulated protection for "users," those users were hospitals, financial companies, and public institutions that use AI systems, not people affected by AI.
The country's human rights commission has criticised the enforcement decree for lacking clear definitions of high-impact AI, noting that those most likely to suffer rights violations remain in regulatory blind spots. Experts say South Korea chose a different path from other jurisdictions, opting for a more flexible, principles-based framework that prioritises "trust-based promotion and regulation."
The effectiveness of this approach remains to be seen, but the global community is watching closely as countries grapple with rapidly advancing technologies.
. Like, 53% of global deepfake victims are in one country? That's wild. But I guess when you're dealing with something like that, you've got to do something. The new law is a good start, but it feels like a Band-Aid solution at best.
. Sounds like they're trying to strike a balance between giving companies some leeway and protecting citizens. But what happens when those two worlds collide? And how do we even define what's considered "high-impact AI"? It sounds like a minefield just waiting to happen 
.
. But hey, only time will tell if this law actually makes a difference or just becomes another bureaucratic headache 
.
It feels like they're playing catch-up rather than getting ahead of the game. And what about those foreign firms that don't have to comply? Are they just gonna swoop in and dominate the market while Korean companies are stuck following the rules?
The law's focus on "trust-based promotion and regulation" sounds like a nice phrase, but is it really doing anything concrete to protect citizens? I mean, if 53% of all global deepfake pornography victims are South Korean, shouldn't that be a major priority? 
Until then, I'm not convinced this law is gonna make a significant difference in the long run. 
But at the same time, their profits depend on it 
I mean, 98% of AI startups were unprepared for compliance and that's a big problem 
! I'm interesting in how South Korea is taking AI regulation super seriously and trying to set a new standard. But at the same time, I'm like, why not cover more citizens? The whole deepfake thing is wild and I feel bad for those who got affected. The law seems pretty good on paper but what's gonna happen when it comes to actual enforcement? 
. And what's up with not providing clear definitions of "high-impact" AI? That's just asking to create more loopholes 
. Still, I guess it's better than nothing 
.
I'm totally stoked about South Korea taking bold steps on AI regulations!!! 
. They're probably gonna get fined big time!
. It's like, AI regulation is the new cybersecurity 
And now we're trying to regulate this stuff, but I'm still not sure if our approach is the right one. 
