AlexNet: The Deep Learning Big Bang
A neural network crushes an image recognition contest by 10 points, igniting the modern AI revolution overnight.
Explore this event on the interactive timeline →In 2012, Alex Krizhevsky, Ilya Sutskever, and Geoffrey Hinton entered AlexNet into the ImageNet competition and won with a top-5 error rate of 15.3%—crushing the runner-up at 26.2%. This 10-percentage-point gap shocked the entire AI community. Within months, every major tech company pivoted to deep learning. Google acquired DeepMind. Facebook hired Yann LeCun. The modern AI era began.
Key Numbers
- Error Rate
- 15.3% (vs. 26.2%)
- Parameters
- 60 Million
- Hardware
- 2x NVIDIA GTX 580
Verified Facts
- AlexNet achieved 15.3% top-5 error vs. 26.2% for the runner-up—a gap so large it was considered impossible by conventional methods.
- Three factors converged: massive data (ImageNet, curated by Fei-Fei Li), GPU computing (NVIDIA CUDA), and algorithmic innovations (ReLU activation, dropout regularization).
- Ilya Sutskever, the student who co-built AlexNet, would later co-found OpenAI and lead the development of GPT-3 and GPT-4.
- AlexNet triggered the acquisitions that shaped AI: Google bought DeepMind (2014, $500M), Facebook created FAIR, Baidu hired Andrew Ng.