European AI Alliance

Archived

ARCHIVED PLATFORM

This website is archived since mid-May 2021 and it is not possible to login anymore.

But don't worry there is a new Futurium platform and all active communities have been migrated.

Forum

Explain the contradiction between politics' AI praise - and reality

Bernd Brincken

17 March 2020 - updated 1 year ago
31 comments
Total votes: 0

Statements like "AI is transforming societies and economies" (OECD) could only be legitimate for Strong AI - systems or species on eye level with Homo sapiens.

Such kind of systems are nowhere near to be seen, nor systems with lesser abilities on the path to strong AI.
Looking at history of science, the AI field is notorious for failing in its own predictions, starting 1956 at the Dartmouth Conference, and several times thereafter.

Still, AI is currently praised to be a crucial technology for the future society - this time not only by science and fiction authors but by leading politicians like Germany's economy minister Altmaier or French president Macron.

How can we understand this contradiction?
What political, sociological or economical mechanism could explain it?

Tags:

Sociology

Comments

Laurentiu VASILIU,

25/03/2020 15:44

Politics is supporting a vision of AI technologies potential, not actual performance/capabilities of AI. There is a wide agrement that is has potential of growth.

What we see as a 'hype' is a relatively normal cycle in any development, be it technology, product or start-tup - that is repetitive. There is always a hype, followed by a 'winter'/reality wake-up and then maturity/stability level.

Finally, AI is a scientific domain to start with, relatively new, made of over 150 different technologies branches, applied mathematical models and formalisms and not one abstract artificial entity.

Bernd Brincken,

02/04/2020 18:22

"There is always a hype, followed by a 'winter'/reality wake-up and then maturity/stability level."

Yes, but the other hypes typically start with something New :-)
Instead of digging out the same idea that was buried 10 years ago, and presenting it as a fascinating future possibility, again and again.

The argument is often that new hardware - which really does appear - could facilitate strong AI now. But this argument is also repeated in every AI hype cycle.

Is there any other hype with a similar pattern?

Laurentiu VASILIU,

03/04/2020 18:02

It is not only about the hardware capabilities that evolved massively, but also about the programming languages (remember COBOL, FORTRAN, LISP, BASIC vs these days' Java, C, Python?) and also the Internet technologies/infrastructure that can be used now in many domains such robotics, health etc. and of course in AI applications for health and robotics. Such applications would have been impossible 20-30 years ago, price-wise and complexity-wise.

'New' can mean many things, including 'invention' and 'innovation. These days mostly 'innovations' happen and are incremental; so AI technologies are improving incremental with various inovations of hardware or software but also with improved mathematical models. Machine Learning and Neural Networks are a good example of evolution in the last 10 years thanks to new hardware/software and theoretical improvements.

Similar hypes are in the medical field - for example oncology: chemiotheraphy and radiation were hailed in the past as *the* tools to cure cancers, and they are standard therapies today; however in spite of their very high financial costs for the patients, their results are modest, hard to quantify and anticipate, while the mortality rate is pretty high. Each new cancer treatment raises maybe too many hopes but too few results for the patiens. And still, no-one is discounting incremental progress in oncology...

Bernd Brincken,

03/04/2020 20:27

And do you currently see an 'incremental progress' towards strong AI somewhere?

Why not start with (a simulation of) animal AI, as an intermediate step?
Why should strong AI not occur on an animal level, with less complexity and hardware challenges, before it reaches human abilities?

For example Blatella germanica, see this discussion at scienceforum.net.

Matthieu Vergne,

04/04/2020 16:26

Why not start with (a simulation of) animal AI, as an intermediate step?

Search for "simulation of rat brain" and you will find that we have already started doing that.

And do you currently see an 'incremental progress' towards strong AI somewhere?

[...]

Why should strong AI not occur on an animal level, with less complexity and hardware challenges, before it reaches human abilities?

Once again, nobody here is speaking about strong AI. The citations you refernce are not about strong AI, but about what they call "AI systems", which are what is currently the hype in the AI domain, which means deep learning and related stuff.

Bernd Brincken,

05/04/2020 00:12

"nobody here is speaking about strong AI"

Oh sorry, but I started this thread with a question about 'strong AI'.

If you want to discuss 'AI systems' instead, you may open your own thread.

Matthieu Vergne,

06/04/2020 20:39

No you didn't: your thread, at least the way it is stated, is clearly about "politics' AI praise" and "reality". Then you stated that heir claims are only legitimate if we speak about strong AI, and thus take it as your assumption. Assumption that I already discarded in long and large through our discussions: you are the only one speaking about strong AI.

Politics don't speak about strong AI because they don't even care about that. They care about what they have at hand, which are the AI systems. I already shown you that what EU politicians are thinking about takes roots in the work of the AI HLEG, group who discarded the notion of strong AI as a secondary matter to focus on AI systems.

Reality is about what is currently done in the AI domain, and there is no such thing as strong AI there beside few debates sometimes. Look at the literature produced in the field and you will see that people speaking about strong AI are a minority.

Whether you speak about politician speeches or reality, strong AI is not the topic, it is a secundary matter because more a philosophical debate than something we can get concrete on. If you wan't to speak about strong AI, feel free to do so, but don't make other people say what you want them to say. If even sourced arguments don't move you on your interpretation, then the discussion cannot go on. You are just taking your own interpretation as a fact instead of what it is: a wrong interpretation based on, apparently, some misunderstandings.

Strong AI discussion can be interesting. Confronting politician claims to reality can also be interesting. But these are to unrelated matters: neither of the seconds speak about the first, so it makes no sense debating about why they do.

Bernd Brincken,

07/04/2020 13:55

Well, my speculation:

They would change their language, priorities and decisions - if they had your wisdom.

Laurentiu VASILIU,

06/04/2020 10:06

Let's look at the human history and we may get a better understanding: water, metals, fire, the wheel was known to humanity since the late bronze age. However the steam locomotive was invented only in 1814. So it took over 2000 years to build up the know-how for such a machine and then imagine, design and execute it.

So while the AI (weak and strong) seems sluggish and hyped, the vision of strong AI will not be realized before incremetal or non-incremental developments happens first in multiple technologies, even if the incremental side seems boring and inneficient.

But why be so impatient and harsh with (strong) AI and its vision?? There are only approx 70 years of consistent effort that is really a little, compared with the time took the modern humanity to build a steam locomotive from its basic parts ;) already available.

From a historical time perspective, AI overall is evolving very fast, maybe too fast in comparison with the wisdom evolution of humans ;)

Bernd Brincken,

12/04/2020 12:46

In human history I can not make out any other example of a technology that was predicted in several waves or generations of scientists and did not arrive.

Matthieu Vergne,

04/04/2020 16:18

Hypes don't need to be "about" something new. They can be caused "by" some new events that bring light over something which may be fairly old, actually. Hypes are all about an exponential interest, whatever is its cause and independently of whether the subject of the hype is new or not.

In science, it is often the case that a mathematical theory is established before to find an actual application to it. Just search for "mathematical theories finding applications later" on your favourite search engine. The day we find an application which can be easily understood by a broad population, it could become a hype depending on whatever other events motivates a broad use. Neural networks are such an example, but only an example.

Here is another example I am familiar with and quite illustrative of what I mean.

HTTPS exists since 1994, but it was created for encrypting banking transactions, so HTTP remained the norm. The disclosure from Edward Snowden in 2013 about how HTTP pages were heavily under surveillance from the US authorities launched a hype towards "full HTTPS", facilitated by the creation of Let's Encrypt in 2014 and its full opening in 2016. This last event brought a way for small websites to generate their own trusted certificates in few clicks for free, thus supporting the hype to higher degrees. During the 20 years after the creation of HTTPS, we have passed from 0% to 30% of pages over HTTPS worldwide, but in just 6 years (2014-2020), we have passed from 30% to 80% because of these security events, not because HTTPS was new or awesome or whatever. It was just an already existing (and good) way to react to a new information brought to the public.

Regarding neural networks, what brought them in a new hype was the fact that we got the computational power to actually train them + some interested people like Yann LeCun who were convinced that it was the right time to bring them back in the light. Since that point, we have found many application cases, and these applications are the new stuff that shed light on the already old technology. Like HTTPS, the hype is not about a new technology, but about new motivations to use it.

Bernd Brincken,

05/04/2020 00:25

No, just re-appearing interest in a technology - like HTTPS or neural networks - is not a hype.
See the definition in Cambridge dictionary:
"A situation in which something is advertised and discussed in newspapers, on television, etc. a lot in order to attract everyone's interest."

What is being discussed in this way in the media is an AI (or prediction thereof) which can replace humans or influence societies, through communication.

This is what this thread is about, not about technological progress that is only perceivable by specialists.

Matthieu Vergne,

06/04/2020 20:52

What is being discussed in this way in the media is an AI (or prediction thereof) which can replace humans or influence societies, through communication.

But this is not the basis for strong AI. Strong AI means consciousness. But this is not the matter here. We are not speaking about an AI that can come in the morning at your working place and take your seat because it made a better impression to your employer than you. We are speaking it the same way we speak about tools replacing humans since automation exists, like the printing activities replacing the scribes.

What is true is that there is a fuzzy zone, where a lot of people think about an AI agent like you do, meaning a strong AI. But that is not what politicians speak about, that is what civilians speak about (and expect) due to ambiguous discourses. And this ambiguous discourse is something that is (willingly or not) maintained by marketers and other decisio makers, because ambiguity give margins to operate, in such a way that if it does not go exactly as you wanted, you can just slightly reinterpret it in your advantage to show you are still doing what you said.

The whole point is that either you speak about informed people, in which case neither of them speak about strong AI, or you speak about uninformed people, in which case you are not speaking about the politicians you cite. In any case, the politicians you cite and the strong AI topic are unrelated.

Bernd Brincken,

07/04/2020 14:28

"There is a fuzzy zone, where a lot of people think about an AI agent like you do, meaning a strong AI. But that is not what politicians speak about, that is what civilians speak about (and expect) due to ambiguous discourses."

Sorry, this is simply wrong - see the various sources in my publication on the matter.
Just one example, of personal experience - in may 2018 conference I met german minister of education and research Anja Karliczek at re:publica in Berlin when she promoted a project 'We can WORK it out' - her statements (translated):
"The possibilities of artificial intelligence present us with a huge task. We move between fascination and fear. Will computers soon be smarter than we are? Then who makes the decisions? Who takes responsibility?"
The same tonage and message is heard by minister Altmaier and chancellor Merkel regularly.

"The whole point is that either you speak about informed people, in which case neither of them speak about strong AI"

Would you classify for example Marvin Minsky as 'informed'?
In 1990 I met him at ars electronica in Linz, Austria, and after his lecture asked him backstage if he was serious with predictions of human-level AI soon to arrive:
'Oh yes, I am dead serious !'
He sent this message regularly:
https://www.siliconrepublic.com/machines/marvin-minsky-ai-predictions

So who of us is living in a 'fuzzy zone'?

Matthieu Vergne,

08/04/2020 23:04

Sorry, this is simply wrong - see the various sources in my publication on the matter.

Unfortunately, I don't know German. But if you have an English version to share (or the raw list of sources you speak about, hopefully non German sources), I would be interested.

Regarding your citation of Anja Karliczek, unfortunately I miss the context to properly interpret her sentences. So I cannot do much with just few sentences. Additionally, translation comes with adaptation (I don't know German, but I am French, I know English, and I translate Japanese too). In particular, there is often words (even complete sentences) that can have several interpretations in a language and, once we choose how to represent it in another language, this representation does not share the same set of interpretations. As long as it has the intended interpretation, that is a good adaptation, but that is something to be confirmed. And since I don't know to which extent you may have adapted it, I just cannot build on it. So that citation is not informative to me.

Regarding Marvin Minsky, again I cannot do much with a single citation from him. Yet you are right, here we touch deep inside the informed group. But so informed that it becomes particularly important to ask them deep details about what they mean when they say "intelligence". Because someone who has a lot of knowledge on a topic can be quite specific in his or her choice of words. And with a term as overloaded as "intelligence", surely he had his own, experience-driven definition, which might be quite different from ours.

So there might be stuff to contradict me in what you said but, if there is, it is still well hidden.

Bernd Brincken,

09/04/2020 13:57

Sure, I have a german perspective.
But when it comes to translation, techniques-which-claim-to-include-AI :-) may help:
- Google-translate on 'Work it out' - en
- French version

So you may get your own understanding about the context of the minister's statement. IMHO It is clearly about AI replacing human workers.
BTW, the arguments and language are similar to those about supporting Syrian migrants when they look for new jobs - desirable IMHO. And that I can only take with humour.

Next step, some ultra-simple research, let's do a google-search for:
"intelligence artificielle" Macron peur
> 545.000 results, just 8% less than leaving out 'peur'.

For example Iin this speech Macron uses 'peur' five times, in the sense of something to overcome when valuing AI:
https://www.elysee.fr/emmanuel-macron/2018/03/29/frances-new-national-strategy-for-artificial-intelligence-speech-of-emmanuel-macron.en

Bernd Brincken,

09/04/2020 14:04

"Regarding Marvin Minsky, again I cannot do much with a single citation from him."

You will have difficulty to find any less AI-affirmative predictions from him, over decades, starting at Dartmouth, ending only with his death. The siliconrepublic article linked above is just one example.
And there are dozens of scientists (more or less, Kurzweil not counting) following in this slipstream, albeit mainly from the U.S.

Matthieu Vergne,

10/04/2020 22:38

Once again, "replacing humans" is not necessarily about "strong AI", as we already discussed when speaking about printing, steam engines, and the internet. It is also not because we speak about fears ("peur" in French) that we necessarily speak about "strong AI".

Let's take the speech of Macron that you cites here (and who says several times "peur" to say that we should *not* be afraid). He speaks about AI as innovation. He speaks about research and economy. He speaks about algorithms and open data. He speaks about AI champions and competition, politics, ethics. He speaks about the initiation of the 3IA project, a national project for AI, for which 4 cities have been selected few months ago, including the one I work in (Sophia Antipolis) and that I follow closely. I regularly attend seminars from companies participating in the project (less since the COVID-19 paid us a visit). He speaks about how to finance AI, how to keep AI talents in France, how to create opportunities in AI, how to make AI researchers and entrepreneurs work together. He speaks about privacy issues, European regulations. He speaks about AI used to predict, support, scale up, relate, etc.

He speaks about a lot of things, but not a single second in 1h of speech is about strong AI. Moreover, regarding the replacement of humans, he clearly states at 33:33:

L'intelligence artificielle ne se substituera pas, ni aux médecins, ni aux chercheurs, mais elle permet ce passage à l'échelle.

My translation:

Artificial intelligence won't replace, neither physicians nor researchers, but it allows to scale up.

Not only strong AI is absent, but your "human replacement" claims are clearly contradicted here. But let's go even further at 48:15:

Mais une intelligence artificielle n’est rien d’autre que le projet qu’elle sert, dans les conditions définies par ceux qui l’ont programmée, sur la base des données qui lui ont été fournies.

My translation:

But an artificial intelligence is nothing more than the project it supports, within the conditions defined by those who programmed it, based on the database provided to it.

Once again, no fear, no consciousness, AI is a tool that works based on how we use it. And during the 49th minute, he makes the parallel with past "innovation" or "grand revolutions", which brought challenges we are familiar with. Again, it is not about strong AI, but about disruptive technologies as we saw in the past, and as I mentionned at the beginning of our discussion.

At 51:50, he says that we will have to put transparence and loyalty in the system (French: il nous faudra mettre à la fois de la transparence et de la loyauté dans le système), make the algorithms public (rendre les algorithmes publics), and other things showing that AI is nothing more than a system we use.

The only times where Macron speaks about what we should be careful, they are not about the machines themselves, but about the other human users, public or private entities, who may use AI systems with different values.

So thank you for this citation, which illustrates very well my position. But I don't need you to help me support my position. What I need is something that illustrates your own position, as clearly as this one illustrates mine. And I won't rely on automatic translation because I am fairly aware of how approximative and unrealiable it can be. And since EU decisions are shared primarily in English, secundarily in all the languages in EU, I have no doubt that, if your claims are supported, there should have something in English already. But so far, I didn't see any. So I am counting on you for that.

In the case where nothing can be found at this level, I assume that, at best, your claims only applies to some German politicians within Germany, and thus this is not here, a EU forum, that you should write your questions, but on a German website to contact German politicians.

Bernd Brincken,

12/04/2020 12:38

Sure, I will provide native englisch sources as well, but just one simple hint before, based on plausible reasoning, what humans can do as well as expert systems:
Of course no politician will ever say "We should be afraid of (x)".
It is his job to send the other message "We can work it out" - by pure chance the title of the german minister's event (BTW, there is an original english site of the re:publica conference available, with lectures like 'Will Robots and Automation Destroy or Save Us?').

But why is it even neccessary for Macron to explain that we should not be afraid?

Let's open an extra thread for this, because on the x-level of indention in this forum, the discussion becomes difficult to read:
https://ec.europa.eu/futurium/en/european-ai-alliance/why-politcs-say-dont-fear-ai

Matthieu Vergne,

13/05/2020 00:46

After a long time, I am finally able to login again.

So let's switch to the other thread you created, but before I will just relativise one of your last claim here: yes there is politicians who literally build their politics on fears, like far-right politicians who feed the fear of foreigners because they will take our job, replace us, all of that. Well, basically more or less the same thing we say about AI now. So yes, there is politicians like that too.

Matthieu Vergne,

18/03/2020 23:57

Considering your basis:

Statements like "AI is transforming societies and economies" (OECD) could only be legitimate for Strong AI - systems or species on eye level with Homo sapiens.

Do you mean that Internet or steam engines or printing didn't change societies and economics when they appeared? As far as I remember history, this transforming property is usually the way we describe these technologies, yet they have nothing to do with strong AI or with any human-like capacity. They are mere technologies to be used, and their use bring significant changes, which has motivated to describe them as such.

At least, when I read this sentence, I interpret it the same way I would interpret it for these technologies. So I fail to see what is your point.

Bernd Brincken,

19/03/2020 12:59

Changing and transforming are quite different claims.
Change happens permanently in all areas of science and society.
Transformation means that several (interconnected) social sub-systems are affected to form the new society.

You are saying that non-strong, tool-like AI (-named) software could have this effect?
Sorry, by my research, the narrative told to policy-makers by 'expert groups' is that the prospective AI could replace human workers - because it has similar or better capabilities.

Let's hear the sentences following the citation from the OECD report:
"[AI] promises to generate productivity gains, improve well-being and help address global challenges, such as climate change, resource scarcity and health crises. Yet, as AI applications are adopted around the world, their use can raise questions and challenges related to human values, fairness, human determination, privacy, safety and accountability, among others."

Could this refer to non-strong-AI, in the sense of steam engines or Internet? Did steam engines or the Internet challenge human values, fairness or accountability?
Sorry, this prediction, or threat perception, only makes sense under the assumption that AI could replace humans, which would be strong AI.
Then, we have no evidence of strong AI, it is pure speculation, 64 years since Dartmouth, at least.

The question of this thread: How can we explain the motives for this speculation?
And this question does have a real basis, other than the ones about AI, because the phenomenon in question - the speculation - is directly in front of us, so we can collect data about it, like in this forum.
Any hints? Any theories?

Matthieu Vergne,

20/03/2020 00:27

Well, let's take the Internet for example :
- promises to generate productivity gains: Internet improved the flow of information, which allows more informed and reactive decisions to be taken, look at the e-commerce, e-learning, or trading activities for example.
- improve well-being: surely you may find a lot of people who think that Youtube and other websites participate to their well being, but you may also consider visioconferences and e-mails or social media, which help to keep in contact and socialize, an important aspect of well being.
- help address global challenges, such as climate change, resource scarcity and health crises: for the example of the Coronavirus, the accessible information helps to inform ourselves in almost real time about the evolution of the pandemic, about the possible actions to do to preserve ourselves, about how to help, etc.
- can raise questions and challenges related to human values, fairness, human determination, privacy, safety and accountability, among others: Internet has raised them all, be it the darkest aspects of humans that can be shown and seen anonymously over the Web (rapes, pedopornography, massacre, etc.), fairness which is hurt by the capacity of people to better exploit the internet to help them take the lead, the influence on human determination possible through social media and their algorithms, privacy issues due to hacking, safety issues as a potential consequence of the leaks of private information, and many more.

So yes, all of that applies to Internet, which I claim to not be a strong AI, otherwise it would contradict the general consensus that we didn't make one yet.

And when I read the report of the AI HLEG for an ethical AI:
https://ec.europa.eu/futurium/en/ai-alliance-consultation

the interpretation of AI as a mere set of tools seems totally applicable to that report. But probably the most robust evidence to that is the dedicated report about the definition of AI used by the AI HLEG group, which you can find below the ethical AI report. This document make it clear, at least for someone familiar with the AI domain, that the group considers current technologies and what we can already expect from them, because they all relate to improvements and issues already observed in the news. The report explains in details, in 9 pages, the different aspects of AI systems to be considered, and strong AI is only a peripheral matter quickly mentioned in page 7. Here is all what is said about strong AI within these 9 pages:

Narrow (or weak) and general (or strong) AI. A general AI system is intended to be a system that can perform most activities that humans can do. Narrow AI systems are instead systems that can perform one or few specific tasks. Currently deployed AI systems are examples of narrow AI. In the early days of AI, researchers used a different terminology (weak and strong AI). There are still many open ethical, scientific and technological challenges to build the capabilities that would be needed to achieve general AI, such as common sense reasoning, self-awareness, and the ability of the machine to define its own purpose.

In brief, it looks to me that your interpretation of AI in your citations as strong AI is:
- not necessary, since I showed above that these sentences clearly apply to current technologies, no need for strong AI to make sense out of them,
- not motivated, since the people writing these documents and the people building on their recommendations clearly start from the current state of the art, which is explicitly considered to be about narrow AI.

Bernd Brincken,

20/03/2020 11:58

Do we agree, that - narrow or not - AI and the Internet are separate fields?

Or is your lengthy explanation meant to show that the Internet is somehow a manifestation of AI?

If so, why is the organisation behind this Forum called 'European AI Alliance' and not 'European Internet Alliance' or maybe 'European Information Technology Alliance'?

The question of this thread - which you did not care to address yet - is about the motives for putting 'Artificial Intelligence' on a silver tablet in the public.
If AI is just one of many aspects of the internet revolution, this question becomes even more relevant.

Matthieu Vergne,

21/03/2020 14:09

Do we agree, that - narrow or not - AI and the Internet are separate fields?

Yes.

Or is your lengthy explanation meant to show that the Internet is somehow a manifestation of AI?

Clearly not. Rather the opposite. I want to highlight that what you reduce to a "strong AI" only statement is actually a generic statement that can be applied to many different things, like marketers, sellers, or politicians are used to say. This is all a matter of presentation to cover many interests in AI.

The question of this thread - which you did not care to address yet - is about the motives for putting 'Artificial Intelligence' on a silver tablet in the public.

Because those who do that have an interest to do so, be it:

economical, like high tech companies who want to sell their products,
marketing, to show the public we are "at the state of the art",
activism, to advertise about one's position for or against AI,
and whatever interest that would motivate lobbying about AI.

This is why I don't get your point, which seems to build on the hypothesis that these statements assume that they can only be about strong AI, which I claim is wrong based on the explanations and documents I mentioned. What you seem to consider to be specific to strong AI is actually a generic speech that could be applied to completely different things. So what motivates it is not any specific reason about strong AI but many different reasons that various people could put in it. And governments are just surfing on the hype because they also have interests in doing so, whatever they are.

Bernd Brincken,

22/03/2020 20:09

Fine, so we agree on Hype.

Then let's talk about the interests you mentioned.
This forum on ec.europa.eu is part of the European Commission which is a political entity (albeit not public, behind an approval curtain). It is not a private company (marketing), not a civil society organisation (activism), not a lobby organisation:

"The European Commission (EC) is the executive branch of the European Union, responsible for proposing legislation, implementing decisions, upholding the EU treaties and managing the day-to-day business of the EU."

This means the part of AI enthusiam (for diplomacy's sake) that can be observed here can only be motivated by economy. Correct conclusion?

So I will open a new thread on the economic perspective of AI, soon.

Matthieu Vergne,

23/03/2020 01:40

Although it may not be a lobbying actor, the influence of lobbying activities is massive in the European institutions. So I would not completely disregard this aspect to motivate the actions and communication of the Commission.

Jesús María SERRANO,

23/03/2020 10:44

In my personal opinion, "AI is transforming societies and economies", is a true statement. Two decades ago, Expert Systems were seen as the spearhead of AI. Today, no one talks about them anymore. Does that mean they're gone? No, that means they have been diluted in other systems and that their influence, although not publicized, is very great. And the rules that govern these built-in Expert Systems are known only to their designers.
AI systems, strong or weak, are following a similar roadmap to Expert Systems and are beginning to make decisions on issues that affect citizens. Fraud systems, credit granting, hospital bed allocation and, ultimately, which people deserve the most attention in coronavirus infection.
I do not want to enter into the discussion of whether this is a transformation or a change, but it is true that it is modifying the decision-making of institutions and companies, decisions which affect the daily development of citizens.

Bernd Brincken,

25/03/2020 13:14

"[Expert systems] have been diluted in other systems and .. their influence, although not publicized, is very great."

Reminds me of my childhood, when I was fascinated by the Apollo program, plus the outlook of an alien civlisation to make first contact to humans in a moon crater, brilliantly illustrated by the british UFO TV series, or the Perry Rhodan novels. A friend of mine threatened our relation when he refused to accept the alien aspect.
20 years later there was a similar dispute with another friend, but changed roles.
My argument was "The aliens might actually exists, but as long as they do not care to give a press conference at the Cologne cathedral, we should not care as well."

Nowadays, theories of the kind "great influence, although not publicized" are classified as conspiracy theory or pseudoscience.

Juan LLORET,

05/04/2020 15:48

Hello, argued and logical debate

We get into discussions of terms, definitions, and sometimes I think one of the problems in dealing with this is the lack of vocabulary (actually used and accepted), taxonomies, and published standards. Fortunately I guess light is opening a few on this. In case it is of interest, I leave some links that should be the tools (in my opinion) to be able to sharpen the ideas. Thanks for the discussion, very interesting.

Juan A.

External links:

Internal link:

https://ec.europa.eu/futurium/en/european-ai-alliance/ai-watch-defining-artificial-intelligence

Bernd Brincken,

07/04/2020 14:34

No objection.

ARCHIVED PLATFORM

This website is archived since mid-May 2021 and it is not possible to login anymore.

But don't worry there is a new Futurium platform and all active communities have been migrated.

Share this:

Comments