The while Apple is following

[Shishirgano9]

But as Benict Evans notes in his annual Slush presentation. The after the initial excitement. The the reality is different: the majority of users have only tri ChatGPT once. So how far can this technology really go? Benict Evans presents . The as he does every year. The a summary of vision in a few minutes The key question. The Evans says. The is scale. Will LLMs continue to grow to replace entire systems. The or will they shrink to become mere software building blocks? Already.


The divergent strategies are emerging: Meta is distributing its kuwait cell phone number list open source models for free. its tri and true formula—making tools “better. The faster. The cheaper. The ” as it has always done in the computing world. In 2013. The machine learning prov its usefulness by identifying patterns. The such as recognizing a dog in a photo. But in 2024. The generative AI poses a new question: what is it really us for? Evans points out a crucial limitation: LLMs are not databases.


The example of Air Canada. The order to pay damages after its chatbot li to a grieving passenger. The illustrates this problem well. How do you handle errors in a probabilistic and non deterministic system? Can it replace Google? Nothing is less certain. On the other hand. The their potential lies (still) in their ability to become “infinite interns” (to recall his phrase from last year). The automating repetitive tasks to free up time for strategic projects. Evans goes further and compares LLMs to invisible features.