人工智能人工智能一种_人工超级智能的最疯狂后果-CSDN博客

人工智能人工智能一种

人工超智能的最疯狂后果是什么？ (What Could Be the Craziest Consequences of Artificial Superintelligence?)

We all, well most of us in any case, remember how Skynet became aware and tried to take over the world. Once it became self-aware, Skynet, with its super artificial intelligence saw humans as a threat and triggered the nuclear holocaust.

我们所有人，无论如何我们所有人，都记得天网是如何意识到并试图占领世界的。一旦具备自我意识，天网便凭借其超级人工智能将人类视为威胁，并引发了核大屠杀。

Maybe we’ve been amazed by how F.R.I.D.A.Y assists Tony Stark in everything that he does. From troubleshooting his Ironman suit to making a reservation at a restaurant, she, or it is always at his side.

也许我们对FRIDAY如何协助Tony Stark所做的一切感到惊讶。从解决他的Ironman西装的问题到在餐厅预订，她还是总是在他身边。

Though we found it entertaining, we never gave it much thought as we just did not have the technology available today. For years AI was only part of good, or bad, science fiction books and movies. The advances in artificial intelligence or AI, however, has certainly made many things we thought to be impossible, possible. If one looks at the current science and technologies available, AI gets real. According to Tim Jones, AI expert at SDS club “Artificial intelligence develops faster than mankind intelligence did”. Who knows what we will face in the nearest decade.

尽管我们发现它很有趣，但是由于我们今天没有可用的技术，所以我们从未考虑过。多年来，人工智能只是科幻小说和电影中好坏的一部分。但是，人工智能或AI的进步无疑使我们认为不可能的事情变得可能。如果看一下当前可用的科学和技术，人工智能就会成为现实。根据SDS俱乐部的 AI专家Tim Jones的说法，“人工智能的发展比人类的智力要快”。谁知道最近十年我们将面对什么。

So, today the question arises, is something we see in the fiction movies even possible? Are machines that could use super intelligent AI a threat to humans? There are some experts which have cautioned against it and warned that it could pose a danger.

因此，今天的问题浮出水面，我们甚至可以在小说电影中看到某些东西吗？可以使用超级人工智能的机器是否会对人类构成威胁？有一些专家警告不要这样做，并警告说它可能构成危险。

Maybe we should take a step back and first look at what artificial superintelligence is. Armed with this knowledge, we can maybe answer the question.

也许我们应该退后一步，首先看看什么是人工超级智能。有了这些知识，我们也许可以回答这个问题。

What is Artificial Superintelligence?

什么是人工超智能？

Artificial intelligence has been with us for a while. Just think of Siri or maybe self-driving cars. Although we don’t give much thought to it, these technologies are a part of our daily lives. Thinking of these examples, AI is easy to define. It is the software-based simulation of human intellectual capacity.

人工智能已经存在了一段时间。试想一下Siri或自动驾驶汽车。尽管我们对此未作过多考虑，但这些技术已成为我们日常生活的一部分。考虑到这些示例，可以轻松定义AI。它是基于软件的人类智力能力的模拟。

Now, what exactly is artificial superintelligence? Like artificial intelligence, it is a simulation of human intellectual capacity, except in this case it far exceeds the capabilities of humans. According to Nick Bostrom, the Swedish philosopher, AI superintelligence is “an intellect that is much smarter than the best human brains in practically every field, including scientific creativity, general wisdom and social skills”.

现在，人工超级智能到底是什么？像人工智能一样，它是人类智力的模拟，但在这种情况下，它远远超出了人类的能力。瑞典哲学家尼克·博斯特罗姆 ( Nick Bostrom )表示，人工智能超级智能“比几乎在每个领域(包括科学创造力，一般智慧和社交技能)都比人类最好的大脑聪明得多 ”。

Although AI hasn’t developed to that level yet, it does make one think what would be possible if that is achieved. It’s no wonder that some leaders in the technology field have warned us against it.

尽管人工智能还没有发展到这一水平，但它确实使人思考如果实现这一目标将有可能实现。难怪有些技术领域的领导者警告过我们。

人工超智能的一些疯狂后果 (Some of The Crazy Consequences of Artificial Superintelligence)

Let’s forget machines taking over the world for one minute, and maybe look at some more tangible negative consequences AI superintelligence can bring. These might not mean the end of the world but the sooner we begin to contemplate what these might be, the better we can equip ourselves against them.

让我们忘记机器占据世界一分钟的时间，也许不妨看看AI超智能可能带来的一些更明显的负面影响。这些可能并不意味着世界的尽头，但我们越早开始考虑这些可能是什么，我们就可以更好地对付它们。

Keep in mind that these might not be all the consequences, but just some of the crazy consequences which can occur.

请记住，这些可能不是全部后果，而只是可能发生的一些疯狂后果。

失业 (Job Losses)

In its current state, some predict a net increase in available jobs due to artificial intelligence. In other words, although AI may lead to some job losses, it will create more jobs. This is as a result of the improvement in automation and autonomous manufacturing. On the other hand, these AI systems need to be created and maintained by humans, which will create jobs.

在目前的状态下，有人预测由于人工智能，可提供的工作净增加。换句话说，尽管AI可能会导致一些工作机会流失，但它将创造更多的工作机会。这是自动化和自主生产得到改善的结果。另一方面，这些AI系统需要由人类创建和维护，这将创造就业机会。

Now take this a step further to a situation where artificial superintelligence is the norm. If the capability of the AI systems far exceeds the intellectual capability of humans, it stands to reason that there will be far more job losses. In such a case the machines will be capable of doing any job better than a human.

现在，将这种方法进一步发展到以人工超级智能为准则的情况。如果AI系统的能力远远超过人类的智力能力，则有理由认为将有更多的工作机会流失。在这种情况下，机器将能够胜过人类。

Imagine feeling ill and going to the doctor, except that the doctor is a machine that does all the tests, looks at your symptoms, and makes a diagnosis. As a bonus, and without you even knowing, it can send the prescription through to the electronic pharmacist who will dispense your medication.

想象一下，您生病了并去看医生，除了医生是一台可以进行所有检查，看一下您的症状并做出诊断的机器。作为奖励，即使您不知道，它也可以将处方发送给电子药剂师，由他来分配您的药物。

Far-fetched? Maybe, maybe not. How close are we to AI? Some experts predict that we can have our first artificial intelligence that can pass as human within the next 40 years.

牵强？也许吧，也许不是。我们离人工智能有多近？一些专家预测，我们将拥有我们的第一个人工智能，它将在未来40年内像人类一样传播。

That brings us to another consequence, a positive one.

这给我们带来了另一种结果，即积极的结果。

医疗护理 (Medical Care)

Humanity has made some great advances in medical and health care. There has been excellent improvement in the care for Cancer, Alzheimer’s, surgery, and other chronic illnesses. If machines with super intelligence and capabilities exceeding those of humans would diagnose, operate, and do research, the possibilities are endless.

人类在医疗保健方面取得了长足的进步。在癌症，阿尔茨海默氏病，外科手术和其他慢性疾病的护理方面已取得了很大的进步。如果具有超人类能力的超级智能和能力的机器能够诊断，操作和进行研究，则可能性是无限的。

This has already been proven in some instances where machine learning algorithms are making more accurate diagnoses compared to human clinicians.

在某些情况下，与人类临床医生相比，机器学习算法可以做出更准确的诊断，这已经得到了证明。

Whilst during a global pandemic, imagine for one moment if we had health care workers who could work around the clock and not get sick. The benefit to the patients would be immeasurable.

在全球大流行期间，请想象一下我们是否有可以全天候工作而不生病的医护人员。给患者带来的好处是无法估量的。

自主武器 (Autonomous Weapons)

For years, nuclear weapons in the wrong hands have been one of the greatest fears of mankind. Luckily it has been kept under control, but the amount of destruction these weapons can cause is reason for fear.

多年来，核武器不当使用一直是人类最大的恐惧之一。幸运的是，它一直受到控制，但这些武器可能造成的破坏程度令人担忧。

Now throw some artificial intelligence into the mix. The thought of autonomous weapons which can’t be easily disarmed in the hands of a government or terrorist organization is frightening. Once deployed, they would not be easy to stop.

现在，将一些人工智能混合在一起。不能轻易在政府或恐怖组织手中解除武装的自动武器的想法令人恐惧。一旦部署，它们将不容易停止。

Even more frightening, is these weapons under the control of a machine that can decide how and against who to use them. Who knows, the next nuclear superpower might be a super intelligent superpower.

更可怕的是，这些武器是由一台可以决定如何使用和针对谁使用的机器控制的。谁知道，下一个核超级大国可能是超级智能超级大国。

骇客 (Hacking)

We hear of hacking attacks every day. From phishing attacks to accounts being hacked and ransomware, it is quite common.

我们每天都听到黑客攻击。从网络钓鱼攻击到被黑客入侵和勒索软件的帐户，这很普遍。

Hackers are already using artificial intelligence in their attacks. They use AI to create and accelerate polymorphic malware. This causes the code to change the code constantly and make it undetectable. They also use it to bypass facial security and spam filters and to bypass anomaly detection engines.

黑客已经在使用人工智能进行攻击。他们使用AI创建并加速多态恶意软件。这会导致代码不断更改代码并使它无法检测。他们还使用它绕过面部安全和垃圾邮件过滤器，并绕过异常检测引擎。

Ultimately, however, these hackers are humans who decide who to attack and why. When we have a machine, which can implement these attacks faster, more efficient and for its own reasons, the results could be mind-boggling. If we take this into account, a Skynet which holds the world at ransom is not such a crazy idea.

但是，归根结底，这些黑客是决定谁攻击和为什么攻击的人。当我们拥有一台机器时，由于其自身的原因，可以更快，更高效地实施这些攻击，其结果可能令人难以置信。如果考虑到这一点，将世界勒索为赎金的天网并不是一个疯狂的想法。

Whether frightening or amazing, the field of artificial intelligence is surely fascinating. Ultimately, whether these consequences occur is up to us. We need to be aware of the risks and prepare for it. One way to do this is to learn more about AI.

无论是令人恐惧的还是令人惊叹的，人工智能领域必定令人着迷。最终，这些后果是否发生取决于我们。我们需要意识到风险并为此做好准备。一种方法是了解有关AI的更多信息。

We can use it for good, some will use it for bad, but it is a part of our future. Who knows, maybe one day we will have a tycoon in a steel suit using his AI superpowers to save humanity.

我们可以用它作好事，有些人可以用它作坏事，但这是我们未来的一部分。谁知道呢，也许有一天我们会用他的AI超能力拯救钢铁大亨。