Many people have probably heard of artificial intelligence but may be unsure exactly what AI entails.

AI occurs in two phases; the learning or training phase, in which case an algorithm is "taught" how to react to incoming information from troves of past data. The second phase is the "inference" phase, in which case a machine reacts to a prompt based on its learning without human interaction. Along the way, there's quite a lot of software, processors, and memory that make all of this work, and there are a lot of companies directly or tangentially involved.

One thing's for sure: The AI revolution is taking off and is bound to make many companies rich in the 2020s. Today, three of the best-positioned AI stocks are CrowdStrike (CRWD 2.03%), Alphabet (GOOG 1.06%) (GOOGL 1.08%), and Lam Research (LRCX -3.27%). Here's why each is a solid buy in September.  

A human finger touches a robot finger with a spark between them and animations emanating from the contact.

Image source: Getty Images.

Using AI to disrupt the cybersecurity market

CrowdStrike (CRWD 2.03%) has more than tripled since its IPO just over one year ago, but that shouldn't deter you from still taking a look at this endpoint cybersecurity company today. Although the company looks quite expensive at a $25 billion market cap and a price-to-sales ratio of 45, CrowdStrike's 85% growth rate and significant gross margin expansion backs up the optimism.

Why is this upstart cybersecurity company on such a tear? Chalk it up to its disruptive business model that was born in the cloud and uses AI to create a "network effect" that continually improves the product based on threat data from all of its customers. CrowdStrike's Falcon platform consists of its easily deployable lightweight agent that can attach to any enterprise endpoint, from servers, to laptops, to mobile devices, to Internet-of-Things devices. Every endpoint sends data back to CrowdStrike's centralized Threat Graph, which amalgamates all of that data to continually improve the company's defense algorithms.

CrowdStrike's impressive management is nothing if not confident. In the company's annual report, it writes, "by analyzing and correlating information across our massive, crowdsourced dataset, we are able to deploy our AI algorithms at cloud-scale and build a more intelligent, effective solution to detect threats and stop breaches that on-premise or single instance cloud products cannot match."

There appears to be something there; in 2019, CrowdStrike catapulted from ninth place in the endpoint security market to fourth place, while the three companies above CrowdStrike all lost market share. What's more, even after last year's big share gains, CrowdStrike had only 5.8% of the overall endpoint security market.

I'd look for CrowdStrike to continue gobbling up more endpoint market share over the years while also entering new segments of cybersecurity in the years ahead. The company reports earnings on Wednesday, Sept. 2, so interested investors may wish to buy a portion of stock now, then see what management has to say after the release and conference call.

Two hands outstretched with an animation of a human head in profile between them.

Image source: Getty Images.

Alphabet is an AI conglomerate

Guess which large company was an early investor in CrowdStrike. That would be Alphabet, which invested in the cybersecurity firm in 2015 through its Google Capital, or "CapitalG" later-stage growth investing entity.

But it's not just through its two investing businesses, CapitalG and Google Ventures, in which Google is exposed to AI. No, AI is at the heart of most core Alphabet products, even its main search engine, where AI-driven algorithms improved the core search function over human-developed algorithms in recent years. YouTube video recommendations also depend on AI, and the rising star in the portfolio, Google Cloud Platform, offers a full suite of AI tools, from image recognition, to voice and language AI, to machine learning and other out-of-the-box algorithms that clients can use. The same goes for Google's hardware products, including its Google Home pod and Pixel smartphones.

Alphabet has continued to heavily invest in AI, establishing in 2017 to work on cutting-edge AI research. Alphabet has even begun making its own customized AI processors called Tensor processing units, which were first developed in 2016. And Alphabet is also making great strikes in futuristic AI-related fields such as quantum computing and self-driving cars with its Waymo subsidiary.

Despite its formidable search business and new high-growth segments like YouTube and Google Cloud, Alphabet has lagged the other FAANG stocks this year, even though it's up 23% for 2020.

GOOG Year to Date Total Returns (Daily) Chart

GOOG Year to Date Total Returns (Daily) data by YCharts

I think investors are focusing too much on the quarter-to-quarter growth numbers in the core search business, which has been temporarily affected by the COVID-19 pandemic and lack of travel-related advertising. Yet given Alphabet's vast cash resources and hefty investments into leading AI research, investors should appreciate the long-term forest, not the short-term trees, making Alphabet a great buy at today's prices.

Two metal prongs etch something onto a round silicon wafer.

Images source: Getty Images.

This semi equipment maker just upped its dividend

One thing is for sure: AI will continue to require advanced processors, graphics chips, memory, and storage. And while many chip companies are chasing that gold rush, the manufacturing of advanced processors and memory only comes by way of a few "picks and shovels" semiconductor equipment companies. Of them, Lam Research is currently reporting some of the strongest profitability metrics. In fact, following its stellar June quarter, Lam Research just announced it was increasing its quarterly dividend by 13%, from $1.15 to $1.30, or $5.20 on an annual basis, good for a yield of 1.5%.

Helping matters is that Lam Research garners an outsize portion of revenues from value-add services relative to peers, at around one-third of its revenue base. Not only do Lam Research's machines make chips for AI applications, the company also uses AI to collect and process vast amounts of data from its large and growing installed base of machines. From that data, Lam has introduced value-add productivity services over the years that help customers reduce defects increase yields, creating a win-win for both Lam and its customers.

Not only should Lam's installed base increase every year as more and more advanced chips are produced, but Lam has been able to increase its "revenue per chamber" by continuing to develop these extra value-add services.

With high profit margins, returns on capital, a steadily growing services segment and a still-reasonable P/E ratio of 23, Lam still looks like one of the best risk-rewards in the tech sector today.