Artificial Intelligence has been a buzzword that has been floating around for decades. Overall for many products and services, AI has been at the forefront of new innovations in how we do things.

人工智能一直是一个流行词，已经流行了数十年。 总体而言，对于许多产品和服务，人工智能在我们的处事方式上一直处于新创新的最前沿。

Most noticeably is within the computer vision space around self-driving cars. There have been huge developments with Musk himself stating that Tesla will have Level 5 autonomous vehicles by 2020 (for a breakdown of what Level 0–5 means see below).

最引人注目的是在自动驾驶汽车周围的计算机视觉空间内。 马斯克本人表示，Tesla到2020年将拥有5级自动驾驶汽车 (有关0-5级的细目分类，请参见下文)，取得了长足发展。

This would be amazing for the industry as well as companies supporting it. Startups like Scale have grown tremendously from the high demand of requirements around helping label and process data for autonomous vehicle companies.

对于行业以及支持它的公司来说，这将是惊人的。 像Scale这样的初创企业已经从围绕帮助自动驾驶汽车公司进行标签和处理数据的需求的高要求中获得了巨大发展。

但是我们看到了2020年的许多变化 (But we’ve seen a lot of changes in 2020)

For technology like self-driving to be successful, you need loads of images to be labeled correctly and fed through the model.

为了使自动驾驶等技术取得成功，您需要正确标记图像负载并通过模型进行输入。

With the wild ride of 2020, conditions across the world have been interesting but also different.

随着2020年的疯狂发展，世界各地的情况虽然有趣，但也有所不同。

With the change in environment brings new sets of data that need to be analyzed and ingested.

随着环境的变化，带来了需要分析和提取的新数据集。

As an example, one clear social norm that has risen that new models need to deal with could be the increased usage of face masks. AI now need to be trained up to correctly identify and categorize faces with masks, especially the upper part of the face.

举例来说，新模型需要处理的一个清晰的社会规范可能是增加使用口罩。 现在需要对AI进行培训，以正确地识别和分类带有蒙版的面部，尤其是面部的上部。

What this all means is that these models still have a long way to go before truly being able to identify what is happening around the world in a constantly changing environment.

这一切意味着，要真正识别出在不断变化的环境中全球正在发生的事情，这些模型还有很长的路要走。

让我们看看另一个例子 (Let’s take a look at another example as well)

Another example could be the massive influx of different search terms in the last few months.

另一个例子可能是最近几个月大量涌入了不同的搜索字词。

In just a few days, terms like phone cases, chargers, LEGO, and more were knocked out and replaced with COVID-19 search terms.

在短短几天内，诸如电话盒，充电器，乐高(LEGO)之类的词语就被淘汰，并被COVID-19搜索词语所取代。

This means a lot of hiccups for AI that run behind the scenes in fields like fraud, marketing, logistics, and more. What is ‘norm’ in 2019 is now completely different in 2020, which means most of these algorithms need to be retrained to keep on top.

这意味着人工智能在欺诈，市场营销，物流等领域的许多幕后动作。 2019年的``标准''现在与2020年的情况完全不同，这意味着大多数这些算法都需要重新培训才能保持领先地位。

“It was never trained on a spike like this, so the system was out of whack,” says Sharma — Global vice president at Pactera Edge, IT Consulting company

“从来没有像这样的高峰训练过它，所以系统变得不合时宜，” Sharma — IT咨询公司Pactera Edge的全球副总裁

Those who simply ‘leave the engines running’ will run into issues, especially those who use AI to help manage inventory and ‘predict sales’ ahead of time.

那些仅仅“让引擎运转 ”的人会遇到问题，尤其是那些使用人工智能帮助提前管理库存并“预测销售”的人。

There are plenty of other examples, and this is just a quick look at how much 2020 has influenced AI.

还有很多其他例子，这只是快速了解2020年对AI的影响。

还有一些较小的问题，例如查找可视数据进行标记。 (There are also smaller problems like finding visual data for labeling.)

AI can help with tasks that could take humans years to do but minutes for an AI to do. For computer vision to work, however, vast amounts of visual data are needed to be processed and labeled (as mentioned before).

人工智能可以帮助完成人类可能需要花费数年才能完成的任务，而人工智能需要花费数分钟才能完成。 但是，要使计算机视觉发挥作用，就需要处理和标记大量的视觉数据(如前所述)。

A lot of this data actually comes from something as manual as someone going into the field and taking a whole bunch of photos. With certain restrictions and lockdowns, this has reduced potential datasets available due to limited access to specific attributes needed for photos.

实际上，许多数据实际上来自于一些手册，就像有人进入该领域并拍摄大量照片一样。 在某些限制和锁定下，由于对照片所需的特定属性的访问受限，因此减少了潜在的可用数据集。

Even worse are areas that are on full lockdown. Some cities, for example, have curfews and limitations of how far you can travel out. Without new visual data, projects will be slower to execute, as well.

更糟糕的是完全处于锁定状态的区域。 例如，某些城市有宵禁和限制您可以外出旅行的距离。 没有新的可视数据，项目的执行速度也会变慢。

一些公司也在降低涉及人工智能的预算 (Some companies are also lowering budgets that are involving AI as well)

There are many companies that use AI for everyday operations. Plenty are backend focused, so you won’t see much of it on the surface.

有许多公司将AI用于日常运营。 很多都是后端关注的，因此您不会在表面上看到太多。

For example, for streaming companies like Netflix, AI is used to help recommend shows and movies that match your taste. This might not seem obvious, but by doing this, it makes you sticky as a customer the more you use their service.

例如，对于像Netflix这样的流媒体公司 ，AI用于帮助推荐符合您口味的节目和电影。 这看起来似乎并不明显，但是通过这样做，您越会使用他们的服务，从而使您成为客户的粘性。

For companies like Google, it is used to help power relevant search results when looking up keywords. Google has had to alter its algorithms due to COVID-19 to ensure the right information goes to the right hands and remove those trying to take advantage of the situation.

对于像Google这样的公司 ，它用于在查找关键字时帮助提供相关的搜索结果。 Google必须根据COVID-19更改其算法，以确保将正确的信息传递给正确的人，并消除试图利用这种情况的人。

但是，并非所有公司都拥有使用AI来帮助优化自己的工作的奢望。 (However, not all companies have the luxury of using AI to help optimize what they do.)

A survey carried out by Boston Consulting Group and MIT polled almost 2,500 leaders and found that seven out of ten said their AI projects had generated little impact so far.

波士顿咨询集团(Boston Consulting Group)和麻省理工学院(MIT)进行的一项调查对近2500位领导人进行了调查，发现十分之七的人表示，到目前为止，他们的AI项目几乎没有产生任何影响。

Two-fifths of those with “significant investments” in AI had yet to report any benefits at all.

在人工智能上有“重大投资”的人中，五分之二根本没有报告任何收益。

With uncertainty around how AI can help produce investment back into a business, there will be a reduction in budgets and projects around it. For example, in a PwC survey, those who said they implemented AI in a ‘multiple of areas’ fell from 27% to 18%.

由于不确定人工智能如何帮助企业重新产生投资，围绕它的预算和项目将会减少。 例如，在普华永道的一项调查中 ，那些说在“多个领域”实施了AI的人从27％下降到18％。

Most of the hype of AI has helped its initial adoption, but with COVID-19, businesses are not only struggling but also needing to cut back investments to enable their survival.

人工智能的大部分炒作已帮助其最初采用，但对于COVID-19，企业不仅在挣扎中，而且还需要削减投资以维持生存。

With something as unproven as AI in a lot of industries, it makes sense there is a significant decrease to these projects and instead a focus on business-critical investments.

由于在许多行业中没有经过AI验证的东西，所以有意义的是，这些项目的数量大大减少了，而是专注于对业务至关重要的投资。

This will most likely continue on as the year goes by until businesses are swinging back to normal and more budget is approved for projects.

随着时间的流逝，这种情况很可能会持续下去，直到企业恢复正常并为项目批准更多的预算。

现在让我们看一下栅栏的另一侧 (Now let’s take a look at the other side of the fence)

Many industries during COVID-19 suffered, but of course, there are winners in all this mess. Companies sitting within healthcare, for example, have seen a boost in sales across the board.

在COVID-19期间，许多行业都遭受了损失，但当然，在所有这些混乱之中都有赢家。 例如，医疗行业的公司已经看到了整体销售的增长。

AI situated within these segments that are booming are also, in return, growing tremendously.

作为回报，位于这些蓬勃发展的细分市场中的AI也在Swift增长。

In a study done by Beauhurst in the UK, there has actually been a surge of demand in 22% of startups across AI.

在英国Beauhurst进行的一项研究中， 整个AI领域的初创企业中，有22％的需求激增 。

In another example, with many companies closing down call centers or cutting down on costs, many have started to rely on AI chatbots to do most of the much-needed interaction required with customers.

在另一个示例中，随着许多公司关闭呼叫中心或削减成本，许多公司开始依赖AI聊天机器人来完成与客户的大多数急需的交互。

现实-人工智能正在挣扎和蓬勃发展 (The reality — AI is struggling and booming)

Unfortunately, as the world battles COVID-19, the reality is, only the industries that can afford to utilize AI within their business will help push the technology further.

不幸的是，当世界与COVID-19作战时，现实是，只有能够负担得起在其业务中利用AI的行业才能帮助进一步推动该技术的发展。

It also depends on what you look at when defining AI.

这也取决于您在定义AI时所看的内容。

There are so many subsets and varying degrees of technology. Some help with mundane tasks, whilst others promise to change the world.

有这么多的子集和不同程度的技术。 一些帮助完成了平凡的任务，而另一些则承诺改变世界。

Depending on the context on either, some are experiencing high demand whilst others will be struggling to find business.

根据两者的具体情况，一些企业的需求量很大，而另一些企业则很难找到生意。