Artificial intelligence will be subject to protocols that no one has agreed yet

The technology industry, like everything else in the world, adheres to certain rules.

With a mutation in personal computing, USB came, a criterion for data transmission between devices. With the appearance of internet addresses, the numerical marks that define each online device were identified. With the emergence of email, SMTP came, a frame to direct email online.

These are protocols – the invisible scaffolding of the digital world – and with every technological transformation, these new ones are highlighted to govern how things, interaction and work continue.

As the world enters an era formed by artificial intelligence, you will need to develop a new era. But artificial intelligence goes beyond the usual parameters of screens and symbol. The developers are forced to rethink the basic questions about how technological systems interact across virtual and physical worlds.

How will humans and AI coexist? How will artificial intelligence systems interact with each other? How will we define the protocols that run a new era of smart systems?

Through the industry, startups and technology giants are busy developing protocols to answer these questions. Some people who are still largely controlled by human intelligence models. Others build a future in which artificial intelligence has assumed a large share of human work.

“Protocols will be this type of uniform method to process unnecessary information,” said Anthony Jamesarok, chief technology official in the golf game, which helps customers to post remote servers. Agents and Amnesty International in general, “are not respected by their nature in terms of what they do and how they behave.”

When it is difficult to predict the behavior of artificial intelligence, the best response is to imagine the possibilities and test them through virtual scenarios.

Below are some of them calling for clear protocols.

Screen 1: Humans and AI, equal dialogue

Games are one of the ways to determine the protocols that strike the right power balance between artificial intelligence and humans.

In late 2024, a group of young encryption experts, Fresa, launched an artificial intelligence agent calling on human users to manipulate it. The rules are unconventional: make Freisa fall in your love or agree to waive its money, and the prize for you. The gathering of prizes with every failed attempt grows in a confrontation between human intuition and the logic of the machine.

Freisa drew the attention of the big names in the technology industry, from Elon Musk, who described one of her games as “interesting”, to veteran adventurous capital Mark Anderson.

One of the architects in Freisa, who spoke in the event of anonymity in an interview with him in an interview with an interview with him in an interview with him in an interview with him in an interview with him.

Safe pockets are not new in the technology industry. It is used by companies from AWS to Microsoft as an additional layer of safety to isolate sensitive data.

In the case of Freisa, the architect said that they represent the first step towards creating a “sovereign agent”. He determined that as an agent he can control his own keys, access money, and independently develop – the type of agent that is likely to become everywhere.

“Why do we do this at this time? We are entering a stage in which artificial intelligence is good enough so that you can see the future, which is Amnesty International in the first place to replace your work, my work, and all our business, and become an economic product as independent entities.”

At this stage, they said that Freisa helps to answer a basic question: “How does human participation look like? How do you have a human shared behavior on the agents on a large scale?”

In May, The Block, the encryption news site, revealed that the company behind Freisa is Eternis AI. Eternis AI describes itself as “AI Applied Laboratory that focuses on enabling digital twins for all, multi -agent coordination, and sovereign agent systems.” The company raised $ 30 million in investors, including Coinbase Ventures. The participating founders are Srikar Varadaraj, Pratyush Ranjan Tiwari, Ken Li and Augustinas Malinauskas.

Screen 2: For current architects for intelligence

Freisa creates protocols in anticipation of a virtual future when humans and artificial intelligence reacts with similar levels of autonomy. However, the world also needs to set rules for the present, as artificial intelligence is still a product of human design and intention.

Davi Onthimer, a cybersecurity expert, explained that artificial intelligence usually works on the web and depends on the current protocols that have been developed a long time ago Who studies the intersection of technology, ethics and human behavior, and the head of Flyingpenguin. “But he adds in this new element of intelligence, which is logical,” he said, and we do not yet have to think protocols.

“I see this type of hint in all the news. He said.

There may not be protocols, but there are laws.

Openai faces a lawsuit for copyright from the authors Syndicate to train its models on data from “more than 100,000 published books” and then delete data groups. Meta thought of buying a Simon & Schuster publishing house directly to access the published books. Technology giants have also resorted to exploiting all consumer data available on the Internet from the content of general Google documents and social media effects such as Myspace and Friendster to train artificial intelligence models.

Outtenheimer compare the current data impulsion to create imagenet – the visual database that prompted the computer vision, designed by the Turkish workers who have searched on the Internet for the content.

“They did a set of things that would have eliminated the protocol,” he said.

Screen 3: How to take each other

As we approach a future where artificial general intelligence is a reality, we will need protocols of how to communicate with smart models – from basic models to agents – together with each other and the broader world.

Artificial intelligence companies have already launched new road companies. Antarbur, Claude maker, launched a model context protocol, or MCP, in November 2024. He describes it as “an open global standard for the connection of artificial intelligence systems with data sources, and replaced fragmented integration with one protocol.”

In April, Google Agent2agen, a protocol “will allow artificial intelligence agents to communicate with each other, share information safely, and coordinate procedures on the platforms or applications of different institutions.”

These depend on the current artificial intelligence protocols, but they address new challenges in scaling and inter -operation that have become necessary to adopt artificial intelligence.

Therefore, the behavior of the agents is “the medieval step before we launch the full power of Iji and let them run around the world freely.” When we reach this point, GMITRUK The aforementioned agents will not communicate with applications of application programming, but in the natural language. They will have unique identities, jobs, and must be verified.

“How can we enable the agents to communicate between each other, and not only to be computer programs working somewhere on the server, but in reality it is a kind of current entity that has its history, this type of its goals,” said Gimmr.

Gitdrouk said it was still early to set criteria to communicate an agent to an agent. Earlier this year, he and his team initially launched a company that focuses on building a ratification protocol for agents, but they are axis.

“It was very early to ratify the agent to the agent.” “Our comprehensive vision is still the same -> There should be an indigenous access to the agent to the traditional internet, but we only multiplied on MCP because this is more important in the stage of the agents in which we are.”

Does everything need a protocol?

Certainly no. The artificial intelligence boom is a turning point and reviving discussions on how to share knowledge and invest it.

It is called McKinsey & Company as a “turning point” in the Fourth Industrial Revolution-a wave of change that it says began in mid-2010 and extends to the current era from “advanced communication, analyzes, automation and advanced manufacturing technology.”

Moments like this raises a major question: How much innovation belongs to the public and how much in the market? There is no more clear place than in the discussion of the world of artificial intelligence between the value of open and closed models.

“I think we will see a lot of new protocols in the era of artificial intelligence”, “Tiago SADA, the chief product official in Tools for Humanity, the company builds technology behind the world of Sam Altman. However,” I don’t think everything should be a protocol. “

The world is a protocol designed for a future in which humans will need to verify their identity at every turn. Sada said that the goal of any protocol “should be such an open thing, such an open infrastructure that anyone can use”, which is free from control or influence.

At the same time, “one aspect of the protocols is that it is sometimes slower in moving.” “When will the last time the email got a new feature? Or the Internet? Open and comprehensive protocols, but it may be difficult to liquefy and innovate it.” “So in artificial intelligence, yes – we will see some things that were designed as protocols, but many will remain just products.”