OpenAI, the maker of ChatGPT officially announced the release of its newest AI model, GPT-4o on Monday.
GPT-4o will now go a step further from just written commands and will be able to converse and connect with images and texts, the announcement from OpenAI said.
“We’re announcing GPT-4o, our new flagship model that can reason across audio, vision and text in real-time,” OpenAI said.
The company said this new version of the language-processing chatbot uses generative AI. It is described as being a step toward a natural, human-computer interaction. It will be able to respond to audio in around half a second, much like a human would.
Similarly, tech titan Google released a showcase series of its AI products on Tuesday.
Alphabet, the parent company of Google unveiled its contender in the AI race across heavyweights like Meta and Microsoft with a new, multimodal AI assistant called Project Astra.
Sundar Pichai, Chief Executive Officer of Google demonstrated Project Astra, powered by a new, upgraded version of the Gemini model at an annual developers’ conference.
In a video demo, the prototype AI assistant responds to verbal commands based on its interpretation of what it sees through a phone camera and a pair of smart glasses.
“Over the past few years, we've been working to improve how our models perceive, reason and converse to make the pace and quality of interaction feel more natural,” Google DeepMind said.
Pichai said Google wants to add these facilities to all 2024 products. Ultimately, the goal is to make Astra easily, but cautiously available.
“The path to productization will be quality-driven,” he said.
The company also made some notable changes to its core search engine. Beginning this week, AI would be able to make independent decisions regarding the tasks and generate searches using verbal and visual commands.
Demis Hassabis, chief executive officer of Google DeepMind, released a statement encapsulating all the new updates.
“By leveraging our leading speech models, we also enhanced how they sound, giving the agents a wider range of intonations,” Hassabis said. “These agents can better understand the context they’re being used in, and respond quickly, in conversation.”
Despite the strides taken by the tech industry towards using generative AI to ease the burden of everyday tasks on humans, the question of consequence remains.
A research paper by the International Association for Public Participation in April 2024 surveyed the potential risks of artificial intelligence.
The paper written by Morgan Boyco, a PhD candidate in the School of Planning at the University of Waterloo, and Pamela Robinson, professor at the School of Urban and Regional Planning at Toronto Metropolitan University, explored the “significant concerns about bias, accuracy and risk” AI tools pose.
“AI is a powerful technology, creating both opportunities and risks of real harm, including ethical and ecological implications, reputational risks and risks to the health of our democracies and how participation is conducted,” the paper said.
Generative AI, despite all the human-like advancements, remains a machine. The integration of AI into daily tasks allows it access to all personal information.
Safety, privacy and security are now everyone’s responsibility. The incorporation of AI in every device may pose a potential threat.
