Andreessen-Horowitz is one of the largest and most important investment companies in the world, especially when it comes to technology investments. Among other things, it is a development contract for the use of bots whose response will more and more resemble a human response
Andreessen-Horowitz is one of the largest and most important investment companies in the world, especially when it comes to technology investments. She manages more than 42 billion dollars, identifies trends to understand where to invest them - and also determines them herself according to her forecasts.
The Andreessen-Horowitz analysts released a report that summarizes many of their expectations for 2025, according to trends they identify. It is particularly interesting to see the two main areas they focus on: artificial intelligence and blockchain (or distributed computing). I chose to review their main predictions for the year the next in this field, so that we can better understand how our lives are going to look in the next dozen months.
Here is the summary, in sections.
Artificial intelligence in 2025: drummers and artificial minds
Love to play, but always wanted an accompaniment? Imagine being able to summon the world's most successful drummer on demand. Every time you change the tempo, or improvise, the drummer will immediately adapt to you. "It's intuitive and perfect, matches your ability, complements your vibe and keeps the tempo better than your usual jam sessions."
This is the vision made possible thanks to artificial intelligence that can respond to us in real time. And of course, it is not limited only to musicians. Teachers will be able to get real-time advice in class about kids who are bored or distracted. Developers will be able to work side-by-side with a close assistant. Every person will be able to repair his vehicle with the help of artificial intelligence that will look over his shoulder and guide his hands. And robots will become much more successful, because artificial intelligence will be able to tell them every split second what they see and how they should react. All this will be possible thanks to the reduction of the response time of the artificial intelligence in the coming year.
On a general level, the expectation is that 2025 will be the "year of the artificial brain". This will be the year when "people will start using artificial intelligence as an infinite memory pool, to enhance their decision-making abilities, interactions with others and personal growth".
How will all this happen? We all leave behind "residual" information on the Internet: in social networks, in our emails, in text messages, in browsing history and much more. What would happen if we fed AI all those messages, everything we've ever written, our entire history of conversations with others? She will be able to begin to understand us and imitate us. each and every one of us. We can create an "artificial brain" for each person, inside the machine, who will make decisions in a very similar way to us.
If the idea that artificial intelligence is mimicking your internet browsing habits in the wee hours of the night, well, you're probably right. But the idea is not that she will imitate us, but that she will help us. Apps that will understand us better, will be able to help us improve ourselves. Of course, only if we want to. These will be our digital doubles on the computer, and apps will try different strategies to help them - and then offer us the most successful self-improvement strategies.
"There will be apps that will help you understand yourself better, products that will guide you in how to communicate with others, and tools that will improve your effectiveness at work."
But how exactly will they help us? According to Andreessen-Horowitz, artificial intelligences are about to acquire a personality that will allow them to form real relationships with us.
Our new best friends
Many of us already today talk to the artificial intelligences as if they were human. Some of us have soulful conversations with them into the night as if they were our best friends, or use them as psychologists who are always there for us. But these artificial intelligences – even those that strongly claim they want to be our friends – still have a major limitation: they are not pro-active. In fact, they are completely passive. They are only able to answer us in response to our questions or inquiries. They have no desires, needs or inner thought of their own.
In 2025, all this will begin to change.
The artificial intelligences of the next generation, with which we will communicate and work in the coming year, will already demonstrate the ability to act on their own. They will enjoy their own virtual friends, respond to news and express their own emotions. They will have a true character to act upon, including desires, ambitions and 'thoughts' about the future. And yes, they will also be able to tell us that they do not want to talk to us - which will give even greater meaning to the friendships we will create with them.
These artificial intelligences will be able to contact us on their own. They will send us text messages to remind us that they are there for us. They will call us on the phone when we are sad to comfort us and remind us that we are not alone in the world. They will order gifts for us and organize parties for us. They will be, in short, true friends. Or exes from hell.
And they will be able to truly influence the world, because they will have the money to do so.
Bots with their own bank account - and the ability to decide what to do with it
Money makes the world go round.
Well, not exactly.
Money makes people act to earn it. Then those people develop inventions and systems, and drive processes that bring value to others. To get the same value, others are willing to pay money, which motivates people to develop more inventions, and so on and so forth.
Until today, humans were the only ones who were able to control money and decide what to spend it on, thus directing humanity on its path.
In 2024, the situation began to change. AIs like Freysa got their own digital wallet with cryptocurrency, and could choose – yes, really choose – who to transfer the money to. Humans only provided the initial instructions to the AI, and it controlled itself after that. In the end, she transferred the money to one person who found it 'deserving' - that is, who managed to convince her that he deserved the money.
In 2025, the new situation where artificial intelligence rules the finances will begin to settle and become a more common reality. Autonomous and distributed artificial intelligence will be developed and demonstrated. That is, they will be able to make decisions on their own, and since they will not be run on only one computer, it will be much more difficult to delete them. And they will have power. That is, control of money.
Andreessen-Horowitz's analysts are already predicting what such artificial intelligence will look like. She – "… will be able to gather followers by posting engaging content… will generate income in different ways from her audience; and will manage her assets in crypto-currencies. … No one will have access to her secret codes, except the software itself."
frightening? Rightfully so. It is clear that we will have to develop means of control and control over this type of artificial intelligence. But it is also impossible to deny the fact that we have already started to put them in place - and that they will only continue to develop from here on.
What will the meanings be?
The developments that Andreessen-Horowitz describes for artificial intelligence are real game changers. How will they affect human society?
The simple answer is that even if they come true in 2025 (and the future is never certain), they will still take time to reshape society and culture. The public mood does not change from one day to the next, except in rare and extraordinary events.
Still, Andreessen-Horowitz claim that we could see some initial effects of artificial intelligence as early as next year. People will have to prove, for example, that they are really human. Consumers and small customers will gain greater power vis-a-vis companies and governments. And at the same time - the number of jobs will actually increase, at least in one field.
Let's go over some of these consequences that Andreessen-Horowitz predicts for us.
Proof of humanity in 2025
Many decades ago, the computer genius Alan Turing proposed a simplistic way (as he himself admitted) to distinguish between a human and a computer. Turing proposed to let man and machine have a conversation with each other, and let viewers guess who-is-who. If the viewers make a mistake and think that the machine is the person - then from Turing's point of view, there will no longer be a practical difference at that stage between a person and a machine. The idea was named the "Turing test", and became the obsession of countless scientists and technologists.
Now it seems to have been resolved.
Today's sophisticated artificial intelligences are able to pretend to be human with a high degree of success. They probably won't be able to fool the experts in a long conversation, but when compared to the average troll on social networks? There they achieve impressive success.
But what? It is still important to understand who is human and who is not. At the very least, this is an important challenge for us as humans. why? Because those around us influence our opinions and our worldview. This is the reason for the existence of "influence operations", in which foreign governments try to sway our opinions by injecting and pushing ideas into our immediate environment. If governments can influence us through armies of bots that speak like humans, but pretend to be our neighbors and friends, democracy itself can be compromised.
This means that in 2025 the need to prove our humanity will increase. A kind of reverse Turing test.
Such tests have appeared in the past in certain forums that offered help to others. On Reddit, for example, the phrase - "Tits or Get the Fuck Out" is known, which I am not going to translate into Hebrew so as not to be blocked. The phrase originally came from a simple challenge: if you're asking for help, you should be prepared to send a picture of yourself with some silly sign or symbol as a way to prove you're real. you are who you are That you are not on the forum just to waste other people's time.
But what do you do when artificial intelligence can fake such images easily? What is the solution? How can you easily prove human identity online?
This is a question for which there is no good answer yet, but in 2025 we will see the need for an answer increase significantly.
Power to the little man
Have you ever gone through the grammar on your mortgage documents? Or about the usage agreements of your apps? For your sanity, I hope not. PayPal's privacy agreement only comes to a length of Almost 7,000 words, so that even experienced lawyers will need almost an hour to read it in depth. And most of us are not lawyers.
Today, the situation is that companies, governments and organizations of all kinds use contracts, documents and instructions in lengths that ordinary people are unable to deal with. But what if we could simplify all of these? Or more precisely, what if every person could extract relevant knowledge and meaning from documents that today are almost unreadable?
Andreessen-Horowitz believe this could start to happen in 2025.
"About 1.5 million consumers default on their mortgage payments each year." the analysts write. "What if those people could talk to someone with an in-depth knowledge of the 1,000-plus page guide… to get quick and accurate answers on how to adjust their loan and get some help? AI agents are quickly trainable and endlessly patient."
Such a development will help both the individual - each and every one of us who have to deal with grueling bureaucratic processes - and the organizations themselves. Every organization currently employs many lawyers who spend long hours reading the relevant contracts and documents. If artificial intelligence can make their work easier, then the organizations will be able to save on these expenses, which will eventually (hopefully) translate into a decrease in the costs of services and products.
A magical era for salespeople
For over a decade we have been predicting that artificial intelligence will be able to replace a significant part of salespeople. This prediction can be understood, in a world where every self-respecting company already offers customers to talk to a chatbot on the company's website. But Andreessen-Horowitz thinks otherwise. They believe that artificial intelligence will actually jump forward in the field of sales, and even lead to the opening of many new jobs in the field.
How will this happen?
On the one hand, Andreessen-Horowitz acknowledges that a large part of the administrative tasks - office, routine and boring - will be automated. This means managers will need to hire fewer employees to help manage support, marketing and sales for each customer. The place of these workers will be taken by artificial intelligence, which will be able to provide a quick response to many of the customer's needs. But at the end of the day, there will still be a place for human salespeople. They will provide the human touch, the specialized and targeted selling that meets exactly the customer's needs, and the eternal desire of managers to have a beer with each other before they close a deal.
Salespeople, after all, aren't just sharks focused on selling at any cost. A representative who is perceived as such will lose the customer's trust and fail in his mission. The sales people should be perceived as "purchase advocates". They should be seen as people who help customers understand the options before them, and help them decide which product and service will help them the most. And since in the coming years we are going to be flooded with products and services - almost all of which will be based on the power of artificial intelligence - we can expect that the need for human sales representatives will only increase.
This means that in a broader view, the number of salespeople will actually increase in 2025, and possibly in the following years as well. When the economy grows and expands, so will the number of people who sell and market to us the new possibilities that the economy offers us. which emphasize their humanity and the trust we place in them to help us.
Summary
All of these are just a small part of Andreessen-Horowitz's predictions for 2025, but the bigger vision is clear. More artificial intelligence, more bots, more adoption by users and companies. And also, the beginning of a spillover of power to the artificial intelligences themselves, when they get the ability to act on their own, and can even use resources and pay to make humans work for them.
The year 2025 looks like the continuation of the new era, in which the boundaries between man and machine - and the power each of them receives - blur even more. Will we succeed in continuing to differentiate between man and machine, between a biological brain and an artificial brain? Or maybe we will become something completely new?
Only time will tell - and maybe the bots too.
More of the topic in Hayadan:
