From DALL-E 2 to ChatGPT, on AI’s wild year | The AI ​​beat

It was my first week at VentureBeat, mid-April. OpenAI had just released the new iteration of its text-to-image generator, DALL-E 2; our lead AI writer, Kyle Wiggers, had moved to TechCrunch before I could choose his brain; and I panicked.

I frantically scrolled through Twitter images of avocado chairs and astronauts riding horses on the moon, wondering what all the fuss was about. I had written about AI trends for over a decade, but it was at a sky-high level — think C-suite tips. Now I was late in realizing how little I understood about the past decade of advances in AI, from machine learning (ML) and computer vision to natural language processing (NLP). Every beat I’ve ever covered has had a learning curve, of course. But the AI ​​beat felt like Mount Everest.

“Give me six months,” I said to anyone who reached out. ‘I’ll know a lot more in six months. For now, I really need you to start from the very beginning.

A year of fast and furious AI development

Nine months and over 120 stories later, I can look back and see that not only was I climbing a steep learning curve, but the pace of news in the AI ​​space was faster than any industry I’d covered before.

Who could keep up with this wild ride? I did what I could: I covered the biggest AI model news, from DALL-E 2 to Google’s Imagen, Meta’s Galactica to ChatGPT. But I completely dropped the ball AlphaFold from DeepMind. I managed to dig deep into DeepMind’s AlphaTensor, with its ability to create faster new matrix multiplication algorithms, but I was too late with Stable Diffusion and its immediate open-source impact.

I tackled the AI ​​laws and regulations: there was the AI ​​recruitment tools under scrutiny, the new AI Bill of Rights and the upcoming EU AI law. There were so many trends to cover, from emotion AI and the ability to cripple AI cyberattacks to deepfakes and MLOps. There were the big brand efforts in AI, including Walmart, Coca-Cola and John Deere, and the big tech news, from Microsoft and Google to Meta and Amazon.

And who could forget the summer focus on AI “sentience”?

Finally, I had the opportunity to interview top AI leaders, including Geoffrey Hinton, Yann LeCun, and Fei-Fei Li, about the 10th anniversary of the so-called deep learning revolution.

It was a lot. But it only made me curious and excited to do more in 2023.

Thank you and a few humble resolutions

So many peers and industry leaders helped me this year as I let my sea legs beat the AI ​​beat.

There are far too many to list, but I definitely want to mention VentureBeat Editor-in-Chief Dan Muse, who I was also lucky enough to work with while at CIO.com, and VentureBeat Founder and CEO Matt Marshall – both gave me a chance to let the AI ​​beat mine.

Thank you to everyone I’ve spoken to on the supplier side, in the academic community, at the big consultancies, at the biggest corporations – you’ve all been patient, supportive and helpful.

To my fellow journalists reporting on AI – including top writers like Will Douglas Heaven, Melissa Heikkilä, Kate Kaye, Will Knight, Cade Metz, Kevin Roose, Khari Johnson and of course Kyle Wiggers – you may not know me, but your reporting has taught me so much. I learn from you every day!

As for my New Year’s resolutions, I humbly promise to do my best to live up to Ben Schneidermans guidelines for journalists and editors about reporting on robots, AI and computers.

I also decide to always have one beginners mindask the best questions I can, and unlike ChatGPT, always admit when I just don’t know.

Happy Holidays and New Year to everyone!

Feel free to reach out at sharon.goldman@businesskinda.com or on Twitter: @sharongoldman