Discussion: View Thread

  • 1.  Suggestion for your "Teams"

    Posted 07-22-2024 12:16 PM
    1. Lose the expression "Soft Skills"
    2. Use the label "Human Skills."

    Cheers,

    Bill



    ------------------------------
    William M. Hayden Jr., Ph.D., P.E., CMQ/OE, F.ASCE
    Buffalo, N.Y.

    "It is never too late to be what you might have been." -- George Eliot 1819 - 1880
    ------------------------------


  • 2.  RE: Suggestion for your "Teams"

    Posted 07-25-2024 10:03 AM
      |   view attached

    Bill, here are something I like to share on human skills – let's say, soft but competent skills. The first is on human-AI team, the second is on cultural differences in the perception of Team Works.

    Human-AI Team

    On this, one can start by imagining a case where the teammate is not another person but an AIPPS (more on Artificial Intelligence). This is already happening in various degrees – in the guise of implementing different AI chips in computers and internet interactions. But in time to come, perhaps on a decadal scale – an AIPPS would likely be one's real teammate (in proper understanding of the term – interactive, collaborative, etc). 

    The 2022 NAP #26355 Human-AI Teaming report made a thorough analyses of where such a Human-AI teaming aspect stands – gaps, research needs, the future, etc. Although their analyses focused on defense establishments, in my opinion, all different engineering communities – would greatly benefit from the authors' analyses and directions. Therefore, thought of sharing it.

    The report findings have been conceived from the perspective of Multi-Domain Operation (MDO) – and here are some shots of them.

    Human-systems integration (HSI): . . . addresses human considerations within the system design and implementation process, with the aim of maximizing total system performance and minimizing total ownership costs . . . HSI incorporates human-centered analyses, models, and evaluations throughout the system lifecycle, starting from early operational concepts through research, design-and-development, and continuing through operations . . .

    AI as a Teammate: A team is an interdependent group of members, each with their own roles and responsibilities, that come together to address a particular goal . . . An AI system can be a member of a team if it takes on roles and responsibilities and can function interdependently . . .

    Characteristics of an Effective Human-AI Team: . . . teams do not begin as effective teams the moment they come together; instead, teams need to train together on individual and team skills. . . Key Characteristics: Team Heterogeneity; Shared Cognition; Communication and Coordination; Social Intelligence.

    Situation Awareness: Situation awareness (SA) is defined as . . . the perception of the elements in the environment within a volume of time and space . . . the comprehension of their meaning . . . and the projection of their status in the near future . . . SA is critical to effective performance. SA has been described as consisting of Four: Situation; Task Environment; Teammate Awareness; Self Awareness.

    AI Transparency and Explainability: Display transparency: Provides a real-time understanding of the actions of the AI system as a part of situation awareness (SA). Explainability: Provides information in a backward-looking manner on the logic, process, factors, or reasoning upon which the system's actions or recommendations are based.

    Cultural Differences in TEAM Perception

    I think I have shared it quite a while ago – likely, during the launching periods of Collaborate ASCE. It is about a study conducted by the University of Montana in1975. American establishments and scholars were curious about Japanese way of doing things – in particular, how it managed to rise up quickly after WWII. Here is a comparison Table (attached), the author came up with.

    I believe, things have changed since the time of this study – in particular, because of enhanced cross-cultural exchanges, infusion and diffusion of ideas and thinking.

    The East (Japan) vs the West (USA) Management Models (from: Japanese Management, Theory Zero not Theory Z, by Masashige Matsuo, 1975. The University of Montana)

    [Attached Table]

    Dilip

    ---------

    Dr. Dilip K Barua, Ph.D

    Website Links and Profile


    Original Message


  • 3.  RE: Suggestion for your "Teams"

    Posted 07-25-2024 01:23 PM

    Always great to learn from you Dilip!

    Re: "here is something I'd like to share on human skills – let's say, soft but competent skills."

    My take is that the education of engineers is absent such knowledge.

    And so predictably at least some 60% of their projects fall short of contractual requirements for scope, schedule and budget.

    Rarely due to tech-issues, but instead due to their inability to know why hearing, listening,

    collaboration, cooperation and communication are EQUAL to their need for tech-excellence.

    Cheers,

    Bill

    p.s.  JUN1995, at ASCE's Education Conference, Denver CO, this need was agreed by the 150+ engineering educators in attendance.



    ------------------------------
    William M. Hayden Jr., Ph.D., P.E., CMQ/OE, F.ASCE
    Buffalo, N.Y.

    "It is never too late to be what you might have been." -- George Eliot 1819 - 1880
    ------------------------------

    Original Message


  • 4.  RE: Suggestion for your "Teams"

    Posted 07-30-2024 09:17 AM

    Greetings,

    It occurs to me that one of the issues in teams is assumption of responsibility. 

    Can an AI team "member" assume responsibility?  I am thinking about times that I as a member of a team expected another member to get something correct and they did not.  Part of being in a team includes working together for accuracy, but AI can produce volumes of information that are hard for a human to fully check, so what happens when a mistake is made by the AI portion of the team and the human portion misses it?



    ------------------------------
    Sarah Halsey P.E., M.ASCE
    New York NY
    ------------------------------

    Original Message


  • 5.  RE: Suggestion for your "Teams"

    Posted 08-01-2024 10:51 AM

    Good question, Sarah. I hope AI would – AI does assume responsibility like an able team member.

    How could AI do it? To answer this question, one has to understand that the function or smartness of an AIPPS – it is as good as the persons behind the machine – as good as they manage programming and training the machine to learn and adapt – in other words, the persons who conceived, developed, trained and brought an AIPPS out into the market for public consumption.

    • As I see it, unlike any other profession – an AI developing team is highly multidisciplinary. There comes the visionary or the philosopher; the programming and modeling scientists and engineers; the persons of the science of human mind or psychologists; the social scientists looking into ethics, morals and laws; and other administrative, financial and marketing supports. Their combined efforts – let's say in the form of the smartness or robustness of an AIPPS – depend on how far the system has managed to graduate from machine learning (ML) to Deep Learning (DL) of machines.

    • Despite all these, in all honesty and realistic expectations, an AIPPS cannot claim to be perfect – so are all models – so are all humans. AI, despite equipped with all different protective shields, is vulnerable to malicious cyberattacks – so are humans in getting influenced, distracted and misled by all sorts of misinformation and disinformation – oftentimes leading to perception and cognition blunders. Therefore, the conformity of AI with reality becomes an issue – and that has led to the rationality of defining a certain level of acceptability.

    • But, like everyone and everything else – an AI developing effort is a continuous learning and adaptive process – a feat of incremental improvements as time passes by. This attitude of learning and adapting is what drives everything – moves the civilization forward.

    • That is where the referred NAP report: 2022 NAP #26355 Human-AI Teaming has something to enlighten us (I have included a short summary in the Artificial Intelligence article). On Limitations, the report touched upon the four: Brittleness; Perceptual limitations; Hidden biases; No model of causation. On AI-Human teaming interactions, it touched upon the five: Automation confusion; Irony of automation; Poor SA and out-of-the-loop performance degradation; Human decision biasing; Degradation of manual skills.

    • AI development processes have come a long way since the fifties – a 7-decade long research and development – from the initial vision of John McCarthy (1927 – 2011) – to a global phenomenon as we see now. It has given birth to the rise of whole new spectrum of human intelligence researches – to program ANN by learning from the BNN processes. The works have come up with different identifications and re-definitions of, e.g. Perceptual Intelligence; Emotional Intelligence; Social Intelligence; and Cognitive Intelligence.

    • Not all AIPPS have the same level of incorporating such intelligence, however. For example, language model, image processing and self-driving auto AIPPS are more impregnated with and tuned to Perceptual Intelligence. Engineering and decision making, on the other hand, require different levels of intelligence – along with the high level, DL of Cognitive Intelligence.

    Dilip

    ---------

    Dr. Dilip K Barua, Ph.D

    Website Links and Profile


    Original Message


Related Content