Demonstrating how fast and far-reaching the artificial intelligence boom is, almost half (47%) of the companies surveyed by CNBC say AI will be their top technology spending priority over the next 12 months, with an AI budget of 2.5%. The answer is more than double. Cloud computing, the second largest spending area in technology, is at 21%.
This is according to the latest survey, conducted twice a year by the CNBC Technology Executive Council, which includes top information officers, chief technology officers, and chief information security officers in the economy, including marketing. , which includes responses from CTOs from companies outside the technology sector. Pharmaceuticals, telecommunications, utilities, companies from the public sector, etc.
Overall, almost two-thirds say their investments in AI are accelerating, making AI investments a bigger part of the overall small pie. Just over half (53%) of tech executives say higher interest rates are slowing overall spending. Still, with AI booming in 2023 and tech stocks leading the market rally, tech companies say they’ve been under pressure to cut costs due to fear since the last survey of TEC members in the second half of 2022. The percentage of executives speaking out has also decreased significantly. The impact of the recession was the biggest technology challenge of the year ahead. This has dropped from over 30% last year to 16% today, with 9% of survey respondents saying meeting customer demand for technology-driven products and solutions is the biggest technology challenge today. from he increased significantly to 26%.
The CNBC survey, which ran from May 15 to June 20, marked the first time the market cap exceeded $1 trillion after NVIDIA expected revenue to surge to $11 billion in the second quarter of fiscal year 2024, the trading period. was carried out targeting There is a growing demand for graphics processors to power artificial intelligence applications like Google, Microsoft and ChatGPT maker his OpenAI.
Lawrence Dutton Eplus | Getty Images
“It’s hard to think of an area where this didn’t help,” said Diogo Lau, chief information and digital officer at Eli Lilly.
He said Lilly is already using generative AI to create patient safety reports and clinical narratives, which will eventually play a role in drug discovery. “What excites me is whether machines can come up with things that humans couldn’t have imagined, such as new molecules for pharmaceuticals,” Lau said.
One of the most anticipated uses of generative AI is customer relationship management, and it’s happening in more and more companies. Eddie Fox, chief technology officer of telecommunications company Mettel, said the company’s care center has built in AI capabilities to read customer emails, interpret their intent, and take action. This will greatly improve the productivity and efficiency of individual employees, and serve customers faster, he said. “It had a huge impact on our incident-related operations, giving our team about 380 additional hours a month[to actually focus on care],” Fox said.
Others at TEC said they are using generative AI to de-bias job descriptions, create marketing images, and manage social media and IT and HR tickets. It is also seen as a tool to bring professional information to younger employees more quickly. Other companies point out that they are in the early stages of deploying code generation tools using generative AI, as well as “co-pilots” of AI across roles and using generative AI to support investment decisions. bottom.
Some explained that their efforts were still tentative. “We’re just in the learning and exploring phase,” said Nicole Coughlin, chief information officer for the town of Cary, North Carolina, a hub for tech startups such as Fortnite maker Epic Games. rice field.
Even as companies across the economy invest more in AI, many of their strategic technology goals cannot be met with the cloud computing infrastructure already built and still being strengthened. Cloud computing remains the most important technology area for most companies, with 63% of TEC members stating that the cloud is very important to their technology strategy over the next 12 months, but AI is the leading answer. 58% of those who Cybersecurity also remains a major threat, with 42% of respondents saying ransomware is a bigger concern today than it was a year ago.
The latest advances in AI are being applied to challenging cybersecurity environments. Jim Richberg, Fortinet’s vice president of information security and field chief information security officer, said his company is working not only to improve large-scale (billions of nodes) generative AI models, but also to improve subsets of them. He said he has been using AI for more than a decade to identify of the model that produces most of the predictive power. “When you look at trillions of data cumulatively, a lot of the accuracy comes from a fraction of the data,” he said.
Today’s data volumes and complex relationships make cyber defense difficult to manage and customize. “Most organizations either react when problems become noticeable enough to be noticed, or rely on implementing best practices or implementing required practices. It could potentially enable posture,” he said.
One of the reasons AI needs to be more widely deployed in cybersecurity is that it is already being used by hackers and could gain an early advantage. Generative AI, at least in the short term, will increase the ability of malicious actors to create socially engineered content that will be difficult for users to distinguish from legitimate data, Richberg said. A malicious attacker could not only steal the victim’s address book, but also her email traffic, thereby focusing on the contents of the victim’s recent conversations with each contact and identifying their respective contacts. Using language and syntax allows you to spear her phishing her message.
“Emails to your boss and emails to your mother will talk about different things and use different language and tone. “It becomes difficult to distinguish between the Similarly, generative AI makes it difficult to distinguish voice and even video facsimiles from legitimate facsimiles. “Most people have already found it difficult to apply the same cognitive filters that can be used to evaluate emails to real-time interactions. It will make it difficult to rely on user training to avoid breaches,” Richberg said. He said.
Joe Levy, technology group president at cybersecurity firm Sophos, said the company has been developing large-scale language models and deep learning AI for years and now uses AI to detect and detect malicious software. It said it can now stop business email compromise (BEC) emails and phishing. Detect ransomware attack attempts, predict and disrupt early ransomware attacks.
“The most interesting thing about this new generation of generative AI is its ability to make every employee in the company more efficient at work,” he said. This includes more effective “threat hunting,” but extends beyond core cybersecurity work to improving customer service responses and reviewing legal contracts. “Technological advances have always helped organizations scale, but they have never really diminished or enhanced human intelligence. Intelligence that can work together on a wide range of tasks.”Levi said.
Many tech executives take the position that AI will not destroy more jobs than it creates. In the TEC survey, 47% of respondents said they believe AI technology will create more jobs than it will destroy. But a further 26% said it would destroy more jobs than it created, and another 26% said it was too early to know.
“We constantly underestimate the social impact of new technologies. Can you think of a technology that hasn’t changed how people interact? It was a real interaction,” Lau said. But he’s more concerned about companies shying away from AI out of fear. “We may need to launch a fund that shorts companies that ban ChatGPT and goes long companies that encourage it,” he said.
Levy said he had good reason to worry even more this time. “While there is generations of historical precedent for the advancement of technology and its positive and negative impact on society, there is no real precedent for technology that is effectively alien intelligence. , there are many things we cannot predict, and that doesn’t mean we should panic, but to protect against AI’s hallucinogenic abilities, new intellectual property concerns, and the inevitable “duality” This means that care must be taken when considering future legislative measures, etc. We will look at the abuse of use and its impact on future workforce composition,” he said.
But he added, “It’s more cautious optimism than worry.”
