作為一個撲克迷，我深知世上沒有“必贏”一說。然而，如果我們隨機選擇100家科技初創企業，我敢打賭，大多數公司都會大肆宣傳自身業務與人工智能技術結合的多么緊密。

這種人工智能的營銷炒作已是愈演愈烈，這一點我們倒是可以理解。不管是Netflix根據觀眾此前的選擇來預測觀眾可能的節目喜好，還是谷歌根據數千萬網民的點擊對搜索結果進行持續的改善，亦或是具有爭議的人工智能系統，包括亞馬遜善解人意的Alexa，人工智能業已成為我們日常生活的一部分。與此同時，商界正在競相使用人工智能技術來解決一系列問題，只是對于主流消費者來說不怎么顯眼罷了。

競相搭乘新興熱門事物的順風車是人類的天性，而且這種現象以前在其他技術領域也是屢見不鮮，例如大數據、云、軟件即服務、移動、網絡2.0，等等。

但是初創企業對“人工智能”一詞的過度使用——或赤裸裸的濫用——如今尤為猖獗。幾乎每一天，我都會遇到一些企業，它們會在自身的營銷說辭大談特談人工智能，并將自己描述為一家人工智能公司，但卻拿不出真正的人工智能技術作為證據。

例如，這些人工智能公司實際上做的是基礎的數據分析。他們的技術都來源于數據，而且結果都被用于實現特定的目的，例如，根據預設定的規則識別發送營銷信息郵件的最佳時機。

這種根據上下文來整理數據的做法也讓此類公司具有了一定的價值，但這并不是人工智能。它們之間的關鍵區別在于：人工智能系統具有迭代性，分析的數據越多，系統就會變得越智能，越能干，而且越自主化。特斯拉的自動駕駛便是一個例子，它能夠根據其車輛在路上行駛的每一英里來不斷進行完善。真正的人工智能功能會徹底地顛覆市場。

很多軟件即服務和自動化公司也用人工智能來標榜自己，但事實上他們所做的只不過是使用數據分析來編排應用和工作流程。這一技術隨著時間的推移不會變的更加智能，而且也無法達到真正人工智能技術的自主水平。

這些公司錯誤地認為，只要其工作與數據或工作流程相關，都可以被稱之為人工智能。它們還肆無忌憚地使用通常與人工智能相關的“算法”一詞。然而，即便某個系統擁有能夠實現某些功能的算法，但它并不一定就能被稱之為人工智能。

因此，在我們投資高舉人工智能旗幟的公司之前，我們應注意以下幾點：公司的業務是否超出了基礎數據分析的范疇？這些公司是否會產生數據廢氣——公司從感興趣的數據源獲取的大量專屬數據？它們是否使用這一數據來打造日益智能化的系統，并轉而產生自身的數據廢氣？它們是否擁有能夠降低流程中人類干預需求的迭代技術（機器學習或深度學習）。

如果一家公司符合上述所有條件，我們還需注意公司是否擁有：在技術層面上深諳機器學習模型的創始人；將這些模型應用至龐大的數據集的獨特方式；面向這些業務極有可能成功的業務模式。

如果有公司聲稱自己在使用人工智能技術，我們應該向公司負責人詢問以下幾個問題：那些聲稱自己是人工智能技術專家的員工在應對巨大的人工智能挑戰方面是否有經驗，而且相對于競爭對手，他們是否有絕對的優勢？他們是否了解打造自主系統所需要的復雜技術細節？他們是否通過吸引人才在市場上開展競爭？

真正的人工智能技術能夠為現實問題提供突破性的解決方案。如果這些公司在現實當中能夠提供這類解決方案，那么它們再怎么炒作也不為過。（財富中文網）

阿里夫·簡莫哈默德是Lightspeed Venture Partners的合伙人。

譯者：Pessy

審校：夏林

As a fan of poker, I know there’s no such thing as a sure bet. But randomly pick 100 tech startups and I’d confidently wager that the vast majority are t weaving AI heavily into their narrative.

AI hype has become intense, and understandably so. The technology has become part of everyday life, whether it’s Netflix predicting what shows we might like based on previous choices, Google’s search results consistently improving based on millions of people’s clicks, or conversational AI systems like Amazon’s Alexa getting to know you. Meanwhile, virtually invisible to mainstream consumers, the business world is agog over harnessing AI to solve a range of problems.

It’s human nature to latch onto the next big thing, and we’ve seen it many times before with other tech buzzwords of the moment: big data, the cloud, software as a service (SaaS), mobile, Web 2.0—the list goes on and on.

But overuse—or flat-out misuse—of the term “AI” in the startup world right now is especially rampant. Barely a day goes by when I don’t come across a company that is molding its marketing messages to the hype and pitching itself as an AI company, without a genuine AI story to back it up.

In a typical scenario, these artificial artificial intelligence companies are in actuality doing basic data analysis. Their technology sifts through data, and the results are used to drive certain outcomes—say, identifying the best time to send marketing emails based on pre-programmed rules.

Such companies may provide value by making data contextually relevant, but that’s not AI. Here’s the crucial difference: AI systems are iterative—they get smarter with the more data they analyze and become increasingly capable and autonomous as they go. Think of Tesla’s Autopilot improving with every mile that its fleet spends on the road. Authentic AI capability is what enables true market disruption.

A number of SaaS and automation companies out there are positioning themselves under the AI banner, even though all they really do is use data analytics to orchestrate applications and workflows. The technology doesn’t get more intelligent over time, and it never reaches the level of autonomy of bona fide AI.

For these companies, AI incorrectly has become a catchall phrase for anything that has to do with data or workflow. They also tend to liberally throw around “algorithm,” a word often associated with AI. But just because a system has algorithms that drive certain outcomes doesn’t necessarily mean it is AI.

Here’s what we look for before we invest in a company making an AI play: Are they doing more than basic data analysis? Are they creating their own data exhaust—a large trail of proprietary data that they collect from interesting sources? Do they use this data to create systems that constantly get smarter and in turn create their own data exhausts? Do they have iterative technology (machine learning or deep learning) that reduces the need for humans in the loop?

If those boxes can be checked off, we look for the following: founders with deep technical understanding of machine learning models, a unique approach for applying those models to a very large data set, and the strong possibility of a successful business model from all of it.

We should ask the following questions about the heads of companies that claim to use AI: Do the people claiming to be AI experts have experience taking on huge AI challenges, to the point that they have an extreme advantage over competitors? Do they understand the intricate technical details of what it takes to build an autonomous system? Are they attracting the talent to attack the market?

Authentic AI can provide groundbreaking solutions to real problems. The companies that can deliver in the field deserve all of the hype coming their way.

Arif Janmohamed is a partner at Lightspeed Venture Partners.