FW: Do We Need Alignment Between Internet Governance and AI Governance?

[Olivier MJ Crépin-Leblond]

In this CircleID article, Charles Mok writes:

Mr. Chris McElroy’s CircleID article, “Alignment Between Internet Governance and AI Governance,” is thought-provoking. Asking twenty-one questions for readers’ consideration, for their opinions and inputs, really provokes me to think a lot.

As a long-time Internet governance participant and a current analyst of global AI governance trends, these are topics I think about a lot. And I know that to many of us in the Internet governance world, we like to think about how we can leverage what we’ve been through and impose those experiences on AI. But I also know that those in the AI or AI governance world never even thought about Internet governance, and many, if not most, of them have no idea about Internet governance, or have never even heard of ICANN.

Continued on https://circleid.com/posts/do-we-need-alignment-between-internet-governance-and-ai-governance

[Alejandro Pisanty]

Olivier,

thanks for sharing. Long-time friend and highly regarded Charles Mok makes a great contribution with this piece (can we invite him to membership in the Dynamic Coalition?)

I am drafting a response to the original paper and to him pointing to a couple important points that have emerged in our discussions in the previous sessions of the IGF, and more: 

1. There is no central point of control or coordination in AI (not yet at least!) as there is in the DNS; 

2. We should be specific about what we mean by "AI", what subset of that large field we are talking about; 

3. It is conceivable that some problems with AI will welcome a solution that includes multistakeholder governance;

4. We'd better have the problems reasonably well defined and bounded; 

5. This remits us more to the general history of regulation, like for tobacco smoking or the environment: large, rich, powerful private-sector for-profit or otherwise powerful actors will only come to the table at some point in the sequence of denying the problem, denying its importance or reach, denying their responsibility and shifting it to the recipients of harms, fearing the imposition of heteronormative regulation and accept preemptively the meekest form of regulation possible, weakening the regulator, capturing the regulator. For the acceptance step there have to be strong incentives, of which I already mentioned fear; there have also to be incentives to come together and "circle the wagons", against which civil society and the technical community must act decisively, and incentives for the moneyed parties to pay for the mechanisms and institutions (this part is totally missed in NetMundial and NetMundial+10)

What do you feel? This of course is THE hot-button issue for this year's IGF (again, enhanced by the current intense geopolitical events of course - there is an instance of KillSwitch which is likely to hurt the involved population badly going on as I write.) So let's start crafting the sessions, intersessional and at the IGF NOW.

Alejandro Pisanty

--
You received this message because you are subscribed to the Google Groups "coreinternetvalues" group.
To unsubscribe from this group and stop receiving emails from it, send an email to coreinternetval...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/coreinternetvalues/dff6267f-22cc-4436-a5de-0275c83a9974%40gih.com.

--

- - - - - - - - - - - - - - - - - - - - - - - - - - -
     Dr. Alejandro Pisanty
Facultad de Química UNAM
Av. Universidad 3000, 04510 Mexico DF Mexico
+525541444475
Blog: http://pisanty.blogspot.com
LinkedIn: http://www.linkedin.com/in/pisanty
Unete al grupo UNAM en LinkedIn, http://www.linkedin.com/e/gis/22285/4A106C0C8614
Twitter: http://twitter.com/apisanty
---->> Unete a ISOC Mexico, http://www.isoc.org
.  .  .  .  .  .  .  .  .  .  .  .  .  .  .  .

[m_bot...@yahoo.com]

Hi Olivier,

 

Thanks for raising this. Very relevant discussion, for two reasons:

• For internet governance: AI is a game changer for how we use the Internet and how we govern it;
• For AI governance: the lessons to learn from Internet governance are, in my view:
• That the impact is global, across border and jurisdictions – like the Internet
• That governance necessarily includes all stakeholders for their part/responsibility in this – like the Internet

 

It is very different in that developments are much faster than when ARPAnet started and over decades evolved in One Internet that reaches all over the world (even if still 1/3 of the world doesn’t have access). The genie is out of the box – cannot be stopped ... and not knowing how this will evolve requires a focus on principles rather than on technology standards.

 

. I do believe that an underpinning infrastructure – the Internet – can help ensure accountability by ensuring integrity of address (origin and destination) and message (this data package has not been tampered with) – at least that.

 

. I also realise it will be much more difficult to get governments around the world to agree on multilateral matters in a world that is becoming increasingly multipolar – and much less One World (order).

 

Does it affect Core Internet Values as such? Maybe not – and clarity of and adherence to these values may be more important than ever.

 

Maarten

--

[Dr. Alejandro Pisanty Baruch]

Maarten,

I think your last point is key - there is a growing challenge to the Internet's core values, and a review of which are more at risk, and how to uphold them, in the current environment is worth a review and that should be one key subject of our work this year. I've said this almost every year since the Dynamic Coalition exists and it has been a productive approach. Let's try in the coming days to craft a brief overview!

Alejandro Pisanty

---

De: m_botterman via coreinternetvalues <coreinter...@googlegroups.com>
Enviado: miércoles, 4 de marzo de 2026 04:08 a. m.
Para: coreinter...@googlegroups.com
Asunto: RE: [CoreInternetValues] FW: Do We Need Alignment Between Internet Governance and AI Governance?

 

To view this discussion visit https://groups.google.com/d/msgid/coreinternetvalues/039f01dcabbe%24cdd7a530%246986ef90%24%40yahoo.com.

[hadia Elminiawi]

Alejandro, Maarten and all, I fully agree on going ahead with reviewing the Internet core values, and examining which ones are more at risk. I would also like to point out that the core values of the Internet are very different from the emerging core values of AI. The Internet core values focus on connectivity and architecture while AI emerging core values focus on ethics and societal impact. From a governance point of view the Internet goes with the multistakeholder model, while AI is heavily regulated and policy driven.

Kindest regards

Hadia Elminiawi

To view this discussion visit https://groups.google.com/d/msgid/coreinternetvalues/7f08a37f7dae4e43b12020f1b1287335%40unam.mx.

[Christian de Larrinaga]

Wise intervention Alex

It seems too tempting to find a toolbox of spanners in your shed
to try to use it to manage a wildflower meadow.

The human computer interface is not currently well served by the IG
community. OK it thinks about different aspects of computing as
infrastructure rather than the experience by the human.

But AI particularly "ai" embedded in devices and systems covers two very
important human computer attributes - interface and agency - to provide
levels of interactivity with data impossible until very recently. The
bulk of "AI" is actually this lowercase ai today it seems. If there are
resource constraints. These seem to be energy, water and industrial
capacity. No single agency can be established to manage equitable
distribution.

AI (caps) does seem to be emergent. I was interested in a recent
dialogue by network operators who are handing over configurations to
software defined network tools. As the Internet itself becomes
dynamically configurable outside direct human agency. It changes the
characteristic from interactive to system.

As Grace Hopper quiped. "before the war things were simpler. Now we have
systems". I suspect the same observation may come home to bite as the
interactions between systems becomes more visible.

Of course as this develops as AI then the likelyhood this will become
invisible holds an existential question for the notion of "Governance".

Asking questions is good. Asking the right questions pivotal

Anyone had their car say "no" yet? As you infringe on a
Policy/regulatory/insurance rule from on high. Shut down without
consideration that human judgement including to disobey a rule is an
essential feature of human agency?



C

--
Christian de Larrinaga

[Vint Cerf]

Internet, writ large, covers the digital waterfront, including AI, so this is timely. I would say Digital Governance is now on the agenda.

v

To view this discussion visit https://groups.google.com/d/msgid/coreinternetvalues/7f08a37f7dae4e43b12020f1b1287335%40unam.mx.

--

Please send any postal/overnight deliveries to:

Vint Cerf

Google, LLC

1900 Reston Metro Plaza, 16th Floor

Reston, VA 20190

+1 (571) 213 1346

until further notice

[Alejandro Pisanty]

Dear Haidia,

thanks for this thoughtful message. There are indeed serious contrasts between Internet governance and AI governance and the ones you add here are very important.

I only partially agree that the core values for AI governance are as you say "heavily regulated and policy driven." I find this statement true for Europe but not equally so in other regions, and also the landscape in each of them is changing rapidly. The US is moving towards deregulation of innovation and control of AI products and services as a matter of national security in the broadest sense, with a "with me or against me" side-taking approach, and China is not far from that description although through very different mechanisms. There is probably a somewhat different approach emerging in India. Countries like Japan and Singapore are seeking a balance that favors investment, innovation and expansion of services more than Europe. Latin American countries and much of Africa are way behind in producing massively-used services and OTOH producing tons of half-baked regulatory and legislative initiatives. 

So my take together with you would be that the approaches to AI governance at this stage are more heterogeneous than IG ever was - and it *was* heterogeneous 30-40 years ago, but since much of it was built from the technical side first, it was under the radar for many governments and thus could be tested and converged by the year 2000. 

Also in both fields a lot of pruning of issues took place, and in AI must still take place, in order to decide *what even is the subject*, with AI meaning so many different things to different constituencies and generating a host of different opportunities and approaches. 

We must seek the lessons learned in IG and see which translate to AI and which don't. 

Alejandro Pisanty

To view this discussion visit https://groups.google.com/d/msgid/coreinternetvalues/CAFd0P2gdpv%2Br3VEfAm2BiWqJc8oO-SxS%2B99TSuNxa8qEQ8yfMA%40mail.gmail.com.