The next step in generative AI is interactive generative AI.

The next step in generative AI is interactive generative AI.

The interactive AI means that the person just talks to AI and then AI follows the orders. The ability to use the natural language makes it easier to communicate with robots. The human operators can give instructions to robots on what they should do next. 

The interactive AI-based translator makes it possible. People can use their languages in communication whenever they want. The translation program can installed on cell phones. And then people can use it like any other social media program. The communication could happen through central computers where the server program translates words to the target language. 

Or the system can use the morphing network or cloud-based architecture where a cloud of cell phones shares their calculation capacity. Interactive AI can also be a powerful tool for programming and any other things like the R&D process of physical tools. 

The interactive AI can give instructions to programmers. Or it can generate the code from examples. Then the AI asks for the databases' names that want to connect with the program. If AI has access to hard disk, it can search those databases itself, and make the paths. 

There is the possibility that the AI has a multi-tasking user interface. If the names are not  said clearly. The AI can ask to write those filenames. Or it can suggest files that look a little bit about the name that AI heard. The AI can give lists about the found names and then the user can select the right one. 

The next big step in the development of AI is interactive, generative AI. Traditional generative AI just follows the order like "draw me a flower". And then the generative AI just creates that image. Interactive AI discusses with the user, it asks what colors and surfaces the user wants to use in the flower. 

The person can ask to make the yellow flower with crystal leaves and metallic surface and interactive AI just starts to make the flower. During that process, after every stage, it asks is customer happy. Is there something that wanted to change, or is there some hue of color that the user wants to use? 

If the user wants the AI to make a painting, the AI might ask if is there some painting, that the user wants to use as the base. Then the AI asks what parts are the best in the painting, and then it might ask what details the user wants to use in the AI-generated version. 

In the case of the R&D process, the interactive AI is the tool that can ask if is there some airfield where the aircraft must fit. Then it can ask some special variables what are the wanted payload and other abilities that the system must have. If AI creates drones it requires the computer code, that it uses. 

The complicated AI requires powerful computers, and those computers require cooling systems and other kinds of things. Also, the sensors and other things like the drone's purpose determine it's size. The things that the AI also requires are manufacturing tools and available materials. Interactive AI is a tool that discusses with people during the generative process. 

If the interactive AI  drives a car it acts like a second driver. The system might observe a human driver, and if that person is tired or nervous, it can ask what's problem, is and tell that the person needs a break. The interactive AI can take information from the traffic control and adjust the car's speed to a level that it must not stop. The AI can also make a report to authorities if the person seems drunk or somehow angry and request traffic police to stop the driving. 

If a driver makes the mistake that takes the car into the traction. AI can fix that error. When AI takes control the clutch separates the wheel from the power steering so that the driver cannot turn the car. That thing is an important detail in robot cabs. If a drunk person can take a car in control, that thing causes a terrible situation. 

When every single wheel is operated in separate electric engines the AI can adjust their rotation direction and rotation speed. These kinds of systems can installed in hybrid and electric cars. This type of vehicle can use diesel- (or combustion)-electric driving systems. And those systems make them very fast. 

https://www.analyticsinsight.net/interactive-ai-a-step-closer-to-conversational-artificial-intelligence/

AI-based chatbots might have a larger scale and faster influence on the world than nobody expected.

 AI-based chatbots might have a larger scale and faster influence on the world than nobody expected. 

Chat GPT and its competitors have already changed the programming industry. Even their freeware versions can create many things faster than human programmers. Even if the AI-based chatbot cannot dump data straight into a file, it can show the example code. Then the human user can copy-paste that code to the programming editor. Of course, there is some kind of changes like database names and file paths, that the programmer must change. 

But also freeware versions of those AI chatbots can boost the effectiveness of the programming. When people ask which of those chatbots is the best, I must say: decide yourself. Even social media applications have some kind of AI-based chatbot extensions. The AI can create complicated programs very quickly by using public databases. But machine learning makes those systems even more powerful. 

The code above is created by Bing (free version) using the command: "Show me a database connection in C++". The complete code is below this text. And you can see how small changes there must be made. The C++ examples include the same code, but Bing finds it faster than human users. That makes it an effective tool that can improve the effectiveness of programming. 

A learning machine is like a worker who keeps the tools that the person often requires closer than tools that are not needed so often. 

In programming machine learning means that the system will store the modules and code that it needs in the shortcut. The shortcut can be the storage on a hard disk. Or it can be a database with a list of home pages that the system requires very often. There is storage about the user interface models, database connections, and SQL databases that the system normally uses. And that makes it faster. 

When a self-operating AI chatbot makes a program it can change the names of databases and their connections in each program. But the fact is that database connections are always similar. if they are written with the same programming languages. The only thing that makes a difference with those things in different programs is the names of databases. 

If the system keeps a book about codes and libraries, that it used in a certain type of software it could be faster and more effective. The AI-based chatbot works in the cloud. That means the users are using the same software that runs in the same entirety. That means the AI can download libraries and the code that it uses in some software into a "standby" position. 

If the AI must not use the public Internet in all its missions it would be faster and more accurate than any human can be. When we talk about "limited AI! we might mean the programming tool that can create the program by using locally downloaded libraries. In that case, the system uses the internet only for asking code models. Then that system can connect those models with the libraries that it needs. The thing that prevents the AI from making the software is that it doesn't have the required programming libraries.  

The cumulation of information makes the AI more versatile. Sometimes people ask: Can AI-based- chatbots reach awareness? There is the possibility that an AI-based chatbot can hire it's awareness from the users. But the thing that the AI-based chatbot can do is it can create new software at a fast speed. And that thing means the AI-based chatbots are the next-generation programming tools. 

https://techxplore.com/news/2023-09-chatbot-self-awareness.html

*********************************************************************

// Standard C++ includes

#include <iostream>

#include <cstdlib>

// Include the Connector/C++ headers

#include "cppconn/driver.h"

#include "cppconn/exception.h"

#include "cppconn/resultset.h"

#include "cppconn/statement.h"

// Link to the Connector/C++ library

#pragma comment (lib, "mysqlcppconn.lib")

// Specify our connection target and credentials

const std::string server = "localhost";

const std::string username = "root";

const std::string password = "";

int main()

{

    sql::Driver *driver; // Create a pointer to a MySQL driver object

    sql::Connection *dbConn; // Create a pointer to a database connection object

    sql::Statement *stmt; // Create a pointer to a Statement object to hold our SQL commands

    sql::ResultSet *res; // Create a pointer to a ResultSet object to hold the results of any queries we run

    try

    {

        driver = get_driver_instance(); // Get a driver to use to connect to our DBMS

        dbConn = driver->connect(server, username, password); // Connect to the MySQL server

        dbConn->setSchema("test"); // Select the database "test"

        stmt = dbConn->createStatement(); // Create a statement object

        res = stmt->executeQuery("SELECT 'Hello World!' AS _message"); // Execute a simple query

        while (res->next()) // Loop through the result set

        {

            std::cout << res->getString("_message") << std::endl; // Print the result

        }

        delete res; // Delete the result set object

        delete stmt; // Delete the statement object

        delete dbConn; // Delete the connection object

    }

    catch (sql::SQLException &e) // Catch any SQL errors

    {

        std::cerr << e.what() << std::endl; // Print the error message

    }

    return 0;

}