Don't dismiss Gen-AI software

Investing.com – Several high-profile software companies offered weak future outlooks during last quarter's earnings season, but Goldman Sachs has advised investors not to discount the sector because of the hype surrounding Gen-ai.

The influential investment bank highlighted the unusual strength of fourth-quarter 2023 results from several software companies, coinciding with expectations of rate cuts and slowing inflation. Those expectations changed very drastically in the first quarter of 2024.

Oracle (NYSE and Adobe (NASDAQ ) delivered reasonably healthy numbers last week, but companies like Snowflake (NYSE recently forecast disappointing first-quarter product revenue due to expectations that customers would cut back on spending in an uncertain economy.

This news came despite accelerating large-scale growth, raising the question of whether the issues driving weaker demand are cyclical or structural.

Hyperscalers are large cloud service providers, which can provide services such as enterprise-scale computing and storage, and include companies such as Microsoft (NASDAQ ), Alphabet's Google (NASDAQ , and Meta Platforms (NASDAQ ).

The divergence between hyperscalers' acceleration and slowdown in enterprise software has been notable, Goldman Sachs analysts said in a note dated June 3, as hyperscalers have been able to invest considerable amounts of incremental capital to generate revenue from of generative artificial intelligence. (Gen-ai).

“We estimate that incremental capital spending above normal cloud CapEx will be at an annual run rate of between $60 and $80 billion. This has driven estimated incremental revenues related to Gen-ai in the range of $14 billion to $16 billion,” Goldman said.

The global software industry is currently generating $700 billion and is one of, if not the largest, capital spending areas, making it susceptible to a high cost of capital.

There are reasonable arguments that new ai priorities have pushed aside consideration of more mature and well-established spending items, the bank added. Software has always been about productivity and now a cognitive aspect has been added, which as a result changes the value proposition and sales cycles.

This raises the question of whether hardware is finally revolutionizing software, after software dominated for much of the last decade.

To answer this, Goldman noted that technology cycles tend to follow a pattern: At the beginning, the cycle is about building infrastructure. Then there are platforms that enable the best use of that infrastructure, while also providing building blocks for building next-generation applications, and the final stretch is building applications.

“For any technology cycle to be successful, there has to be a killer application,” Goldman said, citing enterprise resource planning in the late 1990s, search and e-commerce during the dot-com bubble, and applications in the cloud after the Great Patriotic War. Financial crisis.”

“In each of these cycles, value creation started with the infrastructure layer and moved to platforms and applications,” Goldman said. “In the case of generative ai, we are still waiting for one or a few great applications that can produce results at scale for different end-user domains, be it front office, middle office or back office.”

While the pace of Gen-ai technology continues to evolve, it has not reached the point where killer applications are obvious today.

One big difference in this generative ai cycle is that while incumbents in the previous client-server cycle were slow to adopt the cloud, application software incumbents in this cycle, whether Salesforce (NYSE:), SAP (ETR :), Oracle. , ServiceNow (NYSE:), Adobe, To sense (NASDAQ:) or Workday (NASDAQ:), are developing Gen-ai products with a greater sense of urgency.

Additionally, data will be a critical element to the success of an ai company, and the data ownership advantage enjoyed by companies like Salesforce, Adobe, Intuit, ServiceNow, and Snowflake cannot be taken too lightly.

Application software companies trade at cyclically depressed valuation multiples, the bank said, and it's a worrying time.

“We believe that in the short term, the cyclical aspect of higher rates, if eased, will cause purchasing cycles to ease and therefore the demand outlook for the group should improve in the coming quarters” said Goldman Sachs.

Companies that take generative ai technology architectures more seriously and are open to deep scrutiny and re-architecting of their platforms will ultimately find success.

After all, a surprising amount of shareholder value creation between 2010 and 2020 came from established software companies like Adobe, Microsoft, Intuit, and Autodesk (NASDAQ:), which adopted technology and changing business models to their surprise. of investors.