OpenAI introduces GPT-4o and a new public image as your BFF

OpenAI introduces GPT-4o and a new public image as your BFF

Plinky-plinky music in the lead-up Open AI Live Stream Announcing the GPT-4o and desktop app was both anxiety-inducing and satisfying, complete with raindrop sounds and a ticking clock. It's a fair representation of the dichotomy between what OpenAI does and how it presents itself to the world.

And, from the choice of presenters to the friendly style — including the soothing female GPT-4o voice, which sounds like a kindly kindergarten teacher — the message of the OpenAI event was, “We're your friends; we're others.” Not like tech companies; let us help you.” All of this seems like a very deliberate choice to position the company as an AI maker you can trust despite (or perhaps in light of) ongoing concerns about it. Copyright infringement, Job changeAnd The dangers of misinformation.

From the ground up, the event was designed to be non-threatening and intimate. For starters, it was led by CTO Meera Murthy instead of CEO Sam Altman. Murthy, the brilliant and handsome chief technology officer, has largely avoided controversy. They visited The Sam Altman Rebellion by the Stepping up as CEO While briefly appointed, but representing a steady voice amid the chaos, Altman maintained support.

Murthy was an excellent choice to lead the event. In his comfy jeans and perfect blowout, he exudes confidence and trust, addressing ethics and safety concerns with a reassuring but vague declaration:

GPT-4o presents new challenges for us when it comes to security, as we are dealing with real-time audio, real-time vision and our team is working hard to figure out how to prevent misuse. How to mitigate We continue to work with various stakeholders from government, media, entertainment, across industries, red-teamers, and civil society on how best to bring these technologies to the world.

This was followed by live demos hosted by Research Leads Mark Chen and Barrett Zoff. Joining Murthy in comfortable chairs surrounded by wood paneling and plants, the trio looked like friends having a relaxed conversation in a natural, organic setting. All of this, of course, comes from the demonstration of a fully synthetic technology capable of recreating the human voice that can talk, emote, sing, and even interrupt. All in real time.

Mashable Light Speed

Speaking of comfy chairs, the whole effect was a stark contrast to your usual Big Tech event. At Google and Apple events, you'll usually find a keynote speaker standing on a wide stage, speaking in hyperbole and absolutes. OpenAI had none of that today. Instead, the event represented the opposite of what we might expect at Tuesday's Google I/O. It was all to say, “We're not like the other guys. You can trust us.”

Chen humanizes himself by admitting that he was nervous. GTP-4o, the inhuman being, guided him through breathing exercises to calm him down. All this was designed to reassure the audience that there was nothing to fear from the new technology, as if the GPT-4o was there to calm us all down.

“When we started OpenAI, our initial vision was that we would build AI and use it to create all kinds of benefits for the world,” Altman said in a statement. Blog post Published after the event. “Instead, it now looks like we'll create AI, and then other people will use it to create all sorts of amazing things that will benefit us all.”

But the dystopian optics of a convincing and animated robot voice were not lost on viewers. The GPT-4o's voice quickly drew comparisons to Scarlett Johansson's role as the voice assistant that Joaquin Phoenix's character falls in love with in the film. his. Altman even seemed to get in on the fun Tweeting “His” during the presentation. So, as he posted his ambitious vision for OpenAI, he was jokingly comparing GPT-4o to a sci-fi technology that stripped away human connection.

OpenAI has been saying one thing and doing another for a while now. Its mission is to “ensure that artificial general intelligence benefits all of humanity,” but the company has been accused of training its AI models on content scraped from the web without credit or compensation. Is. It announced its AI video generator, Sora, as a tool to bring creative visions to life, but it It is not yet revealed what the model was trained on.Although There are many suspects It was stripped of data from YouTube and other videos on the web.

The company continues to release the technology without transparency about how it was created, but throughout, Altman and OpenAI insist they are for regulation and safe deployment of generative AI.

The message is that the public should trust what OpenAI is doing. And today's episode, with its warm wooden tones and lighthearted banter and laughter, illuminated that perspective. Whether we believe it or not is a different story.

About the Author

Leave a Reply