IBM CEO Arvind Krishna believes that programmers will simply be able to do more work. Artificial intelligence will not create more than a third of the code.
During the interview, Krishna shared his thoughts on the prospects of artificial intelligence in programming. In his opinion, it is only a valuable technology, but not a panacea. The IBM CEO disagreed with the recent Dario Amodei, CEO of Anthropic, predictsthat 90% of the code could be written by AI within the next three to six months.
«It seems to me that 20-30% of the code can be written by AI, but not 90%. Are there really simple use cases? Yes, but there are a lot of complex ones where it will be zero» — said Arvind Krishna.
He believes that Artificial intelligence will make programmers more productive and will increase the productivity of their employers, rather than destroying the programming profession, as some critics of AI have predicted.
«If you can produce 30% more code with the same number of people, are you going to write more code or less? History has shown that the most productive company gains market share, so you can produce more products, which allows you to gain more market share».
Not that IBM is interested in portraying artificial intelligence as dangerous. The company sells a number of AI-based products and services, including coding aids. This latest statement contrasts with Krishna’s plans announced in 2023. At that time, he said that IBM planned to suspend hiring for positions that could be replaced by AI.
Krishna compared the arguments about AI replacing workers to early arguments about calculators or Photoshop, which were supposed to replace mathematicians and artists. He acknowledged that there are unresolved issues around intellectual property and AI training, but the end result will be positive, and underestimated.
«It is a tool. If the quality that everyone demonstrates is getting better with these tools, this is true even for the consumer — you now consume better quality».
The IBM CEO predicts that these tools will become cheaper over time. Despite the high power consumption of AI hardware, he believes that eventually this technology will use less than 1% of the energy it uses today, thanks to new methods like those demonstrated by DeepSeek.
«I think DeepSeek has given us a preview that much smaller models can be used. Now the question is, do you really need really big models to start with?»
Despite the proliferation of artificial intelligence, Krishna is not convinced that it will help humanity gain new knowledge. Instead, Krishna believes that quantum computing (a technology in which IBM is investing heavily) will be the key to accelerating scientific discovery.
«Artificial intelligence learns from existing knowledge, literature, graphics, etc. It’s not trying to figure out what’s going to happen… I don’t believe that the current generation of artificial intelligence will lead us to what is called artificial general intelligence… where artificial intelligence can have all the knowledge that is completely reliable and can answer questions other than those that Einstein, Oppenheimer, or all the Nobel Prize winners combined have tried to answer».
Arvind Krishna’s beliefs are different from statements by OpenAI CEO Sam Altmanwho argued that the creation of «superintelligent» AI is quite possible over the next few years and will significantly accelerate innovation.
Source: TechRadar
Spelling error report
The following text will be sent to our editors: