2024-07-10 09:30:22
For now, generative AI (or GenAI) is still far from being optimized and capable of generating quality content. “The hype is here, the revenue is not” financial analysts repeat over and over once more.
In just a few months, generative artificial intelligence has emerged as a disruptive technology. According to McKinsey, it has the potential to generate an additional $200 billion to $340 billion in annual value for the banking sector, and up to $660 billion in the retail and consumer goods sector.
For Sam Altman, CEO of OpenAI (publisher of ChatGPT), AI-powered tools will change the way we approach work and life on dozens of fronts, including business innovation. “It will be the greatest technology that humanity has ever developed,” he said in a interview in 2023.
Maybe. For now, the excitement around AI – and GenAI in particular – has turned into hysteria. All the tech companies and startups have been making endless announcements labeled AI (and it sells even better with “GenAI”…). The main goal: attract investors. But the awakening is starting to be painful.
Despite its impressive advances, the technology is built on shaky foundations. While upstream players, including GPU giants like Nvidia, are enjoying a financial windfall from the exploding demand for compute, downstream developers using AI models are struggling to translate this solution into tangible productivity gains or revenue streams.
Major challenges
The interest of companies in AI and GenAI is no longer in doubt. However, while interest is high, actual adoption remains very low, indicating that these technologies are only in their early stages. Different challenges will have to be overcome.
For practitioners, the key challenge is understanding “where and how” to effectively integrate GenAI. This will necessarily involve conducting business cases, assessing risks, and conducting validation tests before scaling up deployments. This prudent approach will allow companies to leverage GenAI strategically, maximizing benefits while mitigating potential downsides.
Security (and ethics) remains a second challenge for those venturing into GenAI. It will be essential to secure training data, ensuring that it is free of errors and biases. Not to mention the risks of data leaks when using ChatGPT. Recently, the New York Times revealed that OpenAI’s computer network had been hacked… in the first quarter of 2023. The publisher seems more talkative when discussing and teasing the power of its future GPT-5 model than on this security flaw…
Despite these two major challenges, GenAI was initially expected to be a game changer by making powerful AI available to everyone. It is expected to drive dramatic improvements in three areas.
- Customer Support: Smarter, faster interactions with AI-powered customer and employee services.
- Creative work: Generating music, images and videos from simple prompts.
- Developer productivity: Accelerate coding, testing, and automation.
As businesses become more familiar with AI capabilities, the adoption rate will increase from basic applications to more transformative AI-driven processes that might have a significant impact on industry practices and operational efficiency. This is when a killer app may emerge.
1720773649
#Generative #killer #app