How Would A Strong AI Affect The Future Of Game Development?


Share on LinkedIn

The presence of Artificial Intelligence (AI) in the world of sports is not a recent invention. Already in 1949, mathematician and cryptographer Claude Shannon considered a game of chess for a player, in which humans would compete against computers.

In fact, games have been important in the development of artificial intelligence, a testing ground for simulations, pre-prepared environments, and realism tests that form the basis of virtual reality.

For example, in the 1989 game Sim City, players handled complex simulations, and conservative AI was used to emulate something akin to realism, namely, deep human characteristics such as spontaneity. The genre of shooting games has also been invented realistically. In the Total War game of 2000, virtual warriors had the same emotions as humans, just as soldiers on the real-world battlefield.

AI Software Development is particularly valuable in the video game world, as game experience depends exclusively on quality. Neanderthal’s almost exclusive visual experience can be cured (no one complained about ‘Padman realism’). A highly polished visual experience is also fine. However, an almost perfect experience, but not quite, is terrible for the point of disorientation and even repulsion. Game designers call it “disturbing valley “, and here you can see some disturbing examples. Put the AI to work and the game can achieve the required reality to overcome the disturbing valley: this video is realistic Shows a great example of animation.

The Gamesutra Gaming website describes several ways in which artificial intelligence technology contributes to the gaming experience: “Professional games already have artificial intelligence implementations that have proved successful. For example, there is a black and white mask (machine learning), FEAR (context-dependent behaviour) (natural language analysis), spore (simulating life forms through data) … just to name a few. ”

All these discrete techniques are oriented towards two fundamental paths: greater realism in artificial environments and/or more natural interfaces between players and environments. With all of this, there is an evolution that will create a seamless environment: Instead of scripts designed in advance, developers will simply build the general environment and its mechanics, allowing artificial intelligence and IoT App Development to generate customized scenarios and intuitive challenges.

AI Is Now Creating The Next Generation Of The Entire Entertainment Industry

Nowadays, a series of improvements in technology (in consoles, in the cloud and connections, in ultra-powerful graphics cards, in virtual reality and headphones, or in graphic processing algorithms) are improving artificial intelligence and Cloud Computing Solutions, which in turn provides An increasingly influential environment in which virtual characters display human behaviour and intelligence.

Facing two artificial intelligences to compete with each other in games like Go DeepMind led to some of the greatest advances in artificial intelligence (AI) in recent years, since through these games it was shown that machines learn skills through of the method of trial and error that ultimately leads them to overcome humans. But, can the same technique produce more useful artificial intelligence that is capable of operating in the real world?

OpenAI, an artificial intelligence research group based in San Francisco, published an investigation showing that what it claimed is a method to train intelligent systems that are increasingly powerful to which they could prepare to address human problems more ordinary. Set in increasingly realistic environments, the technique points to a path for artificial intelligence to “evolve” in a simulated world until it is ready for use in the real world, he said. The researchers used several intelligent “agents” in a game of hide and seek that was performed in a simulated physical environment. Those who “hide” themselves learned how to cooperate and adapt to their surroundings by moving walls and blocking doors with large blocks to create places to hide. The “seekers” responded, for example, using ramps to jump over the walls.

After 43 million games, the artificial agents had gone through six adaptations while developing response strategies and tactics to be smarter than the other. “As agents learn, they implicitly create new tasks for other agents to solve,” said Bowen, principal investigator of the article. He pointed out “there are many multi-agent problems in the world” that in the end, technology could learn to address, from the design of robots that can operate surrounded by people to the construction of “smart cities” where millions of people interact.

Research in so-called multi-agent problems is not something new, and the work was based on proven reinforcement learning methods, giving artificial intelligence a “reward” every time it manages to solve a problem. But Artificial Intelligence Solutions has setting that the task in a simulated physical world opens the way to “scale” the challenges of the real world, allowing the complexity of the simulation to increase progressively. The biggest challenge, in fact, may not be found in the design of intelligent algorithms, which evolve largely on their own, but in the creation of a simulation that is full proof on a large enough scale. Even in the limited game of hide and seek, agents learned to cheat simulation physics, for example, by finding ways to push the ramps through the outer wall to eliminate them from the game. Unexpected actions like this highlighted the problem of controlling artificial intelligence when it is finally ready to go out into the real world, Baker said. Judges without face or body are already tested In Europe, Estonia announced that it intended to initiate tests with the use of an artificial intelligence agent for the determination of contractual disputes.

According to what was raised by this nation, the two parties in dispute would send the “digital judge” information to support their allegation and the system would issue a verdict, which at least initially, would be reviewed by a human judge to make it a decision. valid judicial. But Estonia is not the only place where virtual judges begin to deal with human affairs; Some artificial intelligence services help countries like the United States and the United Kingdom to appeal traffic fines and some companies that offer AI systems to advise judges in areas such as criminal. However, the Institute of the Right to Telecommunications stressed that these situations open up relevant questions and questions, because if a human judge issues a guilty verdict, there are mechanisms to review the judge’s actions and even reverse his decision.

Amit Agrawal
As the Founder and COO at Cyber Infrastructure Inc, it is my aspiration to drive our global clients ahead in the competitive technology world by enabling them to receive huge financial and operational benefits in web development through my years of experience and extensive expertise as technology adviser and strategist. In my current position at CIS, I spearhead management of various technology initiatives, expansion of our technology capabilities, and delivery of quality excellence to our clients.


Please use comments to add value to the discussion. Maximum one link to an educational blog post or article. We will NOT PUBLISH brief comments like "good post," comments that mainly promote links, or comments with links to companies, products, or services.

Please enter your comment!
Please enter your name here