Human-AI collaboration

LinkedIn post

Konrad Kiss
I'll give a 30 min talk at AI Summit Budapest on Tuesday at the MKIK Prompt Arena about the future of Human-AI collaboration. Let me know if you can make it, I'd love to meet! ❤️

Lorand Kedves
Én nem jutok el ilyenekre, 25 év kutatás-fejlesztés nem kap helyet a nagyok asztalánál. Lásd még: "It is difficult to get a man to understand something when his salary depends upon his not understanding it." (Upton Sinclair, 1935) - egyébként ez érvényes a teljes mai chatbot-AI világra, ahol a néma értő olvasás kizáró feltétel (pl. a Turing teszt definíciója).

A cím alapján viszont kíváncsi vagyok, vajon szóba került-e az alapfogalmak tisztázása során JCR Licklider Man-Computer Symbiosys cikke 1960-ból? 🤔
https://groups.csail.mit.edu/medg/people/psz/Licklider.html

Konrad Kiss
Nem, de örömmel beszélgetnék a témában, ha érdekes lehet Önnek is! Az előadás célja inkább az volt, felhívjuk a kutatásaink eredményeire a figyelmet. Az MKIK segítségével sikerült bejutnunk az eseményre amiért nagyon hálásak vagyunk. :)

Lorand Kedves

Nos, én örömmel beszélgetek ezekről a dolgokról olyanokkal, akiket a tények és az elfelejtett múlt is érdekel, nem csak az aktuális hype meglovagolásából nyerhető pénzügyi nyereség és kapcsolati tőke... 🙃
Sajnálatos módon ezek ellenkező irányba mutatnak, ez az oka a fent említett problémának még a hivatkozott forrásokról tudó környezetekben is. 🤷‍♂️

Ha elfér egy kis önreklám, itt van két (nem túl jó) videó:
https://lnkd.in/duJEbZty
Ezeket egyébként az alábbi előadás bevezetőjéből ollóztam:
https://youtu.be/WKfEDicpwPw

Néhány a vesszőparipáim közül, hátha megtetszik valami 🙂

 

Konrad Kiss

Megvan a szerepe a hypenak és a pénzügyi nyereségnek is. A vállalkozói oldalon ez az életbenmaradást jelentheti, akadémiai oldalon néha ködbe burkolózik. A teremtett érték a kulcs metrika, ez pedig szubjektív. Meg fogom nézni az előadást hamarosan, köszönöm a linkeket.

Lorand Kedves

Ezért jutottam arra a következtetésre, hogy az informatika aranykora 1945-től (MEMEX) 1972-ig (az utolsó Apolló küldetés) tartott, amíg mérnöki és tudományos célokat szolgált: pontos definíciók, objektív értékelés. Ebből persze nem lehet pénzt csinálni, de jött Bill Gates és feltalálta az informatika üzleti modelljét (1976), Alan Kay (tőle nyúlta Jobs) a "végfelhasználót".

2011-ben még csak a következményekkel küzdöttem:

So, if you really solve a problem well, you have solved it for anyone who can run that software, as long as they have the environment that you used. In "business language": if you do your job well in programming, you lose your job.
...
Software industry had invented the "artificial software aging" by changing the environment all the time: creating new operating systems, end supporting "old" hardware, end supporting new hardware with drivers in old operating systems and ensuring that old software will not run in the new environment. New frameworks and generated new requirements on the end user side can only be fulfilled by new and newer software again – so users pay programmers to solve the same problem every year.
https://lnkd.in/dy5Tnm3n

Figyelmeztetés: a háttér megértése jobban fáj. 🤕

 
Lorand Kedves

Az "értékteremtés" kérdéssel kapcsolatban itt ez a dia 👇 amit az (egyébként elutasított) informatika PhD kutatási beszámolómhoz készítettem, de végül kivettem belőle. 2018-ra már megtanultam, hogy semmi értelme olyasmit mondani, amit a "felettem álló" bíráló nem akar hallani.

A három görbe csúcsai olvashatók úgy is, hogy kinek hol az érték (piros: "fogyasztó", kék: "üzlet", zöld: "tudomány"). Tudományon nem a mai publikáció-gyár / pop-science értendő, hanem ami az Apollo programot elvitte a Holdra. Nem a mozivásznon vagy egy VR sisakban (pedig ott "terem a pénz" 🤑 ), hanem a való világban (ami az üzletember és a politikus szerint "csak viszi a pénzt", és mivel "ők tartják fenn az akadémiát" 🤮, onnan is eltűnt 🤷‍♂️ ).

Na ezen a szinten kezd igazán fájni a történet, amikor kezd összeállni a valós viszony ember és intellektus között. Mellesleg, a Turing teszt közismert változatának egy lehetséges fordítása: miért is kéne intelligensnek tekinteni egy olyan fajt, amely épp most vágja el a saját torkát? Miért lenne cél az, hogy a gép ezt utánozza - pláne, milliárdokat és terawattokat borítani a "megoldásába"?
https://www.iea.org/reports/energy-and-ai/energy-demand-from-ai

Kiábrándító.

Outsiders

@lkedves comment
Greetings, Grim, 

I like this model but think it is incomplete and wonder what you think about my version. 

1: The chess analogy is power-oriented, describes actions, a black box model. 

I use a white box model focusing on motivations and found a twisted version of the Indian caste system more useful with the Brahmins (monks), Kshatriyas (guardians), Vaishyas (organisers), Sudras (craftspeople), and the Dalits (untouchables). These represent the five ways that people can deal with their inevitable weakness and mortality; and dictates every decision they make. The model also shows how "modern society" flipped this system upside down, together with a possible (though quite improbable) solution. 

2: There is no single person who can flip the table, this is not how it works. 

There may be people who never accepted the motivations of this game, spent their lives outside of the struggle for money and power, seeking for real answers and better questions(!) They lost all their fights not because they were not ABLE to win but because they were NOBLE to take only what they needed and not what they could grab. They can and do tell everyone their way to play the game but get the response that "you have to understand". They do understand but they don't obey. 

Then the Pharisees went out and laid plans to trap him in his words. They sent their disciples to him along with the Herodians. 

“Teacher,” they said, “we know that you are a man of integrity and that you teach the way of God in accordance with the truth. You aren’t swayed by others, because you pay no attention to who they are. Tell us then, what is your opinion? Is it right to pay the imperial tax to Caesar or not?” 

But Jesus, knowing their evil intent, said, 

“You hypocrites, why are you trying to trap me? Show me the coin used for paying the tax.” They brought him a denarius, and he asked them, “Whose image is this? And whose inscription?” “Caesar’s,” they replied. 

Then he said to them, “So give back to Caesar what is Caesar’s, and to God what is God’s.” 

When they heard this, they were amazed. So they left him and went away. 

Matthew 22:15-22

Today the whole table is in free fall, no pawns and no kings can really control it, the game plays itself through them. Players either realise that and start looking for those outsiders or we die together. 

The outsiders do no look for honour, they seek respect in its original meaning: "re-specto" = "examine it again". Don't reject because you think that's insane. It is just frightening because it is out of the world you knew. 

Your move?

Stargate... Another few billons for Sammy's Toy Factory?

@lkedves1 day ago
25+ years in working with AI (NOT the current chatbot bubble) and the true pioneers of informatics say: it does not matter how far you go into a dead end street because you did not even know that there were maps showing where to go. The question is: when do you realise the mistake and turn back from it... (no problem if you delete this comment, just a minority report)


@rwlurk1 day ago
there is no insatiable demand for AI, but there is insatiable demand from investors for more growth in tech stocks, which are propping up the USA's security markets

@Abouttime-p8u4 hours ago
@lkedves it just might be too lste by then... AI makes work so much easier though, 10 hours of work turn into 1

@Abouttime-p8u4 hours ago
@rwlurk My reply: Have you seen how many billions of pare using AI? Just from the time ChatGPT started to a year later it tremendous growth. Sure, there is also a lot of demand from stock investors interested in companies related to AI.


@lkedves27 minutes ago (edited)
​ @Abouttime-p8u Ironically, you are right but maybe not for the reason you think. Social media is not a big fan of facts and real science, I try to be as lightweight as possible here. Let's quote Alan Turing, highlighting the keys that nobody seem to read. 

--- 

1. The Imitation Game 

I propose to consider the question, "Can machines think?" This should begin with definitions of the meaning of the terms "machine" and "think". The definitions might be framed so as to reflect so far as possible the normal use of the words, but this attitude is dangerous, If the meaning of the words "machine" and "think" are to be found by examining how they are commonly used it is difficult to escape the conclusion that the meaning and the answer to the question, "Can machines think?" is to be sought in a statistical survey such as a Gallup poll. But this is absurd. Instead of attempting such a definition I shall replace the question by another, which is closely related to it and is expressed in relatively unambiguous words. 

The new form of the problem can be described in terms of a game which we call the 'imitation game." 

--- 

Understand? 

Turing stated that both the goals and the means of this current "mainstream AI" are absurd. The "imitation game" does not answer the question if machines can think but replaces it. Turing wrote this article in the Mind (a Quarterly Review on Psychology and Philosophy !) to explain why people should stop calling the Universal Turing Machine a "thinking machine". I don't think he could imagine that a whole industry will be based on misinterpreting the first page (and completely ignoring the next 21) Or rather, just quoting Benedict Cumberbatch from the movie because he is so cute. Scientists... 

I have spent good 25 years on finding a definition of these terms on par with the UTM. I follow Douglas Engelbart, a mentioned pioneer when read AI as Augmenting Intellect, because his forgotten results hold the key. My daily job is to deliver solutions to "impossible missions" in real life with real deadlines and restricted budgets, not hype riding. 

The ratio is not 10 to 1 but 1 to 0. 

On the other hand, social media defines "thinking" as skimming over the popular news following the latest hypes (including videos like this), conduct some excited discussions with relatable buddies in your idea bubble and regurgitate some word salad as your well-founded opinion. And you are absolutely right. 

LLMs can do this much better than 10 to 1. 

But, there is a question. Is this second path really worth billions of dollars and terawatts of consumption? 

Back in the days before LinkedIn turned to a GeekBook, I picked up a quote that still holds today: 

I suppose one way for a machine to pass the Turing test is to wait until the quality of actual human conversation is so bad that a bot could be an improvement. This seems to be happening here. 

Peace... 

---- Interestingly, the first comment remained public and got a response to it ----

@rhettmelton

I've been thinking about this in addition to a "maximum wage" in the USA. Thinking of ceilings for progress. I'm becoming ever-more convinced that we should impose ceilings as a society, to slow the rate of progress for the sake of security and happiness. 

EACC folks (effective accelerationists) encourage us to barrel forward because AGI will solve all of our problems. I have extreme doubts. We will see how it okays out.

@lkedves

Quite the opposite. If you understand that our monetary system is based on exponential functions (the definition of the yearly interest on your bank deposit, or the GDP), combine it with enough elementary mathematics or informatics to know what it means, you realise that there is no ceiling. The repeated hype-rides and falls are simple consequences of the systemic ignorance, under various, sometimes bittersweet ideologies. 

I often watch The Big Short, you only have to replace the few words and get the AGI story - the two mortgage brokers are just like Sammy boy and the whoever that showed the site to Emily. I play a mix of Mark Baum and Michael Burry. Informatics offered a solution but the middle men took over, check Joseph Weizenbaum, the creator of the first chatbot, ELIZA, 1966. For contrast, Softbank invested in all hypes so far: Theranos, FTX, WeWork... I would rather consider that a red flag. Let alone, a praise from Trump.

I wonder if Bloomberg shadow bans this comment individually, or I am banned as a person. Testing, testing... 

😉

What Is The Future Of Programming Languages?

The questions are excellent, the answers are "state of the art", the latter is not a compliment in this case. Here is a different take on the graph part.

0. You have two fundamentally different ways to transfer and curate knowledge, A: storytelling (very human, imprecise) or B: knowledge graph building (hard for a human, as precise as can be). 👉 JCR Licklider, Libraries of the Future (1965, book).
1. STEM knowledge is always B: a graph. When you have a problem in physics, biology, math, medicine, ... it's NOT about how you sing it or what language you use, but to build a precise network of property pages filled with data and linked to each other. The very terms (labels of data and links) are also graphs (DSLs). Information systems are graphs, too. In the computer's memory, you have flowcharts of the algorithms and you use the memory to hold the content of those property sheets. 👉 Ivan Sutherland Sketchpad (YouTube)
2. EVERY program is a combination of a DSL set (the "meta" layer of classes, members, functions and their parameters) and a bunch of stories (the code of the functions and procedures), quoting Bob Martin: the software is assignment statements, if statements and while loops 👉 Future of Programming (YouTube).
3. The real problem is that the text-based tools make us focus on the storytelling, we only see the big list of features or use cases, instead of the DSLs that allow us to describe and solve the atomic problems (modularity, KISS, DRY, SRP. ...). We have already solved every possible atomic problems literally millions of times, but repeating them every time (copy-paste or in "modern cases", LLMs🤦‍♂) in the endless possible combinations, and that grows every day. 👉 Alan Kay's Power of Simplicity (YouTube)
4. Introducing new programming languages that do the same has one effect: it even erodes the "burden" of cumulated human experience and the codebase, start it all over again, not solving the fundamental issue.

Problem: Information systems are graphs. Storytelling is not the right way to interact with graphs. The blamed imprecision is manageable "human error" in the case of graphs but inevitable fatal block in text-based programming. Teaser 👉 Bret Victor, Future of Programming (YouTube); hard-core answer 👉 Douglas Engelbart: Augmenting Human Intellect (report)

Solution: STEM languages are graphs (DSLs). THE future programming language is the DSL of information systems. The same that we have in physics, mathematics, biology, ...

Question: has anyone read this far? 😉

---[ discussion under a comment ]---

CallousCoder
Your behaviour is like the old horse and cart people against the automobile. It's nonsensical, the technology is here to stay. So either adopt it or you'll go extinct. You know that good developers are terrible managers, right? ;) Also I don't get it, what the resistance is. Whether you ask a junior or medior to implement something or you ask an LLM. It's no different other than that the LLM just does it and doesn't nag especially after the 2nd or 3rd iteration ;)

CallousCoder
“bro” is 52 years old and didn’t take philosophy but EE and CS.

lkedves
Age is just a number (happens to be the same...) Check the Mother of All Demos, that was real technology behind the Apollo program, while this chatbot AI is just another stock market bubble. Side note: before the previous AI winter, we won Comdex '99 with a data mining / AI tool. Back then people could read the first paragraph of Turing's article, the definition of "the test" instead of trying to implement cartoon dreams... (including a Nobel prize winner psychologist) 

But you got the point with "Modern software is a disease!" LLMs learn from their sources, kids copy-paste them to real software, LLM learns from them again. Quantity goes up, quality goes down. LLM companies use human slaves to avoid stupid mistakes in everyday tasks after the first flops. Regardless of we accept this as a solution, who will censor generated codes? 

Dead end.

cyberfunk3793
AI is obviously going to fix data races and buffer overflows and every other type of bugs you can think of. You don't understand what is coming if you think it's just hype. I don't know if it will be 5 years or 50, but at some point humans will only be describing (in human language) what they want the program to do and reviewing the code that is produced. Currently AI is already extremely helpful but still makes a lot mistakes. These mistakes will get more and more rare and the ability of AI to program will far exceed humans just like computers beat us at chess.

TCMx3
Chess engines did not need AI to curbstomp us at chess. Non-ML based engines with simple table-bases for endings were already some 700 points stronger than the best humans. Sounds like you don't actually know very much about chess engines lmao.

CallousCoder
btw playing chess with an LLM is a hilarious experience. If it looses it brings back pieces from the dead or just “portals” them into safety.

lkedves
[retry, I promise I leave if disappears again]

You may have missed so I repeat. We won Comdex '99 with a data mining / AI tool (and there is nothing new on this field except the exponential growth of the hardware). Since then I have worked on refining knowledge graph management in information systems under every single project I touched, often delivering "impossible missions". I work together with the machine because I follow a different resolution of AI: Augmenting Intellect (Douglas Engelbart) on systems that are of course smarter than me (and generate part of their own codes for years from these graphs). Right now at the national AI lab in a university applied research project that I will not try to explain here.
You find some of my conclusions with references to sources in my comment added to this video (11th May). 

I know the pioneers who predicted and warned about what we have today (recommended reading: Tools For Thought by Howard Rheingold, you find the whole book online). One of them is Alan Turing, who asked people not to call the UTM a "thinking machine" and wrote an article in Mind: A Quarterly Review of Psychology and Philosophy about the dangers of making such claims without proper definitions. Poor man never thought that in a few decades "IT folks" would think this was an aim. Or Joseph Weizenbaum, the guy who wrote the first chatbot, ELIZA. 

I know why your dream will never happen because I know informatics (its original meaning, not the business model Gates invented) was against this fairy tale. LLMs are just try to prove that old story from quantum mechanics, that infinite monkeys in infinite time will surely type in the whole Hamlet. The problem is that we don't have infinite time and resources, and the goal is not repeating but write the next Hamlet. Those who initiated informatics, made this clear. Start with Vannevar Bush: As We May Think, 1945. 

@TCMx3 , @CallousCoder - thanks for your answers... 🙏 Another excellent example - in chess, you have absolute rules.

In life, we know that all the laws we can invent are wrong (Incompleteness theorem) and thinking means improving the rules while solving problems and taking the responsibility of all errors. The ultimate example is the Apollo program with Engelbart's NLS in the background, that's how THEY went to the moon. We go to the plaza to watch the next Marvel story in 4D, now with the help of genAI. If we go with the question of predicting the next 50 years, look up "Charly Gordon Algernon 1968" here on YouTube.

---[ This answer "disappeared" for the second time so I left the place ]---

Immersive Technologies Policy Primer - reaction

On LinkedIn

This looks like a solid overview of the current "state of the art". What I don't see is the background, 𝐭𝐡𝐞 𝐫𝐞𝐚𝐥 𝐠𝐢𝐚𝐧𝐭𝐬 𝐨𝐧 𝐰𝐡𝐨𝐬𝐞 𝐬𝐡𝐨𝐮𝐥𝐝𝐞𝐫𝐬 𝐰𝐞 𝐚𝐥𝐥 𝐚𝐫𝐞 𝐬𝐭𝐚𝐧𝐝𝐢𝐧𝐠 (and were afraid to look down so now we fall like a stone, just as predicted -> https://youtu.be/KZqsWGtdqiA?t=102 ).

When discussing the social effects of communication technologies, where is the reference to Neil Postman? -> neilpostman.org
https://youtu.be/QqxgCoHv_aE

When talking about education, where is PLATO?
https://youtu.be/THoxsBw-UmM

When discussing AR/VR, where is Ivan Sutherland?
https://youtu.be/AFqXGxKsM3w
Or at least, Alan Kay (among others, the real inventor of the tablet)...
https://youtu.be/pUoBSC3uoeo?t=5061

When thinking about informatics in general, where is Douglas Engelbart?
https://youtu.be/O77mweZ8-RQ?t=22
Or Ted Nelson?
https://youtu.be/KdnGPQaICjk

...

---

"Primer"... 🧐

“Those who cannot remember the past are condemned to repeat it.”
George Santayana

---
... and I guess this date should be 1994:
[3] Paul Milgram and Fumio Kishino (1884)

Haiku

The anti social medium. The Matrix dessert.

The crack on Gödel's Incompleteness Theorem.

The ultimate form of non-transformable information.

The Experience of Being Wrong

How could I be THIS STUPID???

Now, please stop reading and remember when you asked this, not just lightly but with the strange mix of real anger and shame.

...

That was the last time you learned something really important and this is the only way to it. It happened to me yesterday. At age 52 I think this is a very positive feedback: I can (very hardly but still) lower my ego and learn. Here is the story.

I have been repeating for years in every context that JCR Licklider separated transferrable and non-transferrable knowledge, and a root cause of today's mess in IT (and consequently, everywhere) is the fact that we forgot this. Banging my chest like a gorilla, like here...

But yesterday, as my young colleague is creating a proper scientific publication, we started looking for the exact reference. To my greatest surprise, I did not find it. In desperation, I started reading Libraries of the Future again, and realized (thankfully and ironically on page 2!) that...

Licklider never wrote that.

Here is the actual quote:

We delimited the scope of the study, almost at the outset, to functions, classes of information, and domains of knowledge in which the items of basic interest are not the print or paper, and not the words and sentences themselves —but the facts, concepts, principles, and ideas that lie behind the visible and tangible aspects of documents. The criterion question for the delimitation was: "Can it be rephrased without significant loss?" Thus we delimited the scope to include only "transformable information." Works of art are clearly beyond that scope, for they suffer even from reproduction. Works of literature are beyond it also, though not as far. Within the scope lie secondary parts of art and literature, most of history, medicine, and law, and almost all of science, technology, and the records of business and government.

He talks about "transformable information", not "transferrable knowledge".

What happened? Had I forgotten to read???

No, but I was not able to at that moment. This paragraph held a key to a question of computerized knowledge management I struggled with for decades, literally. When it hit me, my mind was blown immediately and started restructuring itself. I followed, remembered, and kept quoting my own revelation instead of the text that I thought I was reading. 

But why?

Knowledge in our minds is always a network: some attributes of and relations between "things". To store or transfer our knowledge of a topic, we "export" the related part of this network in a presentation, text, figures, pictures, videos. Other people will try to integrate this content with their existing knowledge. Here comes the trick: for those who can do this without changing anything in their minds, this was not "information" because information is only the part that you did not know and could not figure out from your existing knowledge.

At the first time I could not integrate Licklider's original message with my existing knowledge, it only triggered a change that took a long time. Now, when I revisited this paragraph, it was new again but now I could actually read it and integrate with my current knowledge. Fun fact: the word "respect" does not mean "obey" or "accept" but re-specto: examine it again.

Real information is like good chilli: it burns twice.

So, how do I read the message now?

This is a simple way to tell the difference between transformable and non-transformable information. I quoted the rest correctly: informatics (the "libraries of the future") should work only with transformable information.

• Transformable means you can say it hundreds of ways, the meaning will be the same. You focus on the knowledge graph in your head and try to build exactly the same in the audience: a physical phenomenon or a medical treatment. 
• Non-transformable information focuses on the message itself and the feelings created by it (not less important but totally different). With different tone, wording, or face, the message and the effect significantly changes.

A less nerdy example

I think 99% of modern pop music is not even information: repeats the same message about a boy, girl, love, hate, etc. that the audience is already familiar. (hashtag metoo?)

But the Sound of Silence is a perfect example of non-transformable information: I already knew the original song but this presentation by Disturbed delivered the message (which happens to be in close relation with the topic of this post).

[This post can be a pair of my more formal article, The Science of Being Wrong (a possible definition of informatics) as here I defined "infonauts" as experts in being wrong...]