Welcome to the forefront of conversational AI as we explore the fascinating world of AI chatbots in our dedicated blog series. Discover the latest advancements, applications, and strategies that propel the evolution of chatbot technology. From enhancing customer interactions to streamlining business processes, these articles delve into the innovative ways artificial intelligence is shaping the landscape of automated conversational agents. Whether you’re a business owner, developer, or simply intrigued by the future of interactive technology, join us on this journey to unravel the transformative power and endless possibilities of AI chatbots.
Make no mistake! Sam Altman is closely following up on his “code red” that he shared with colleagues last December amidst a resurgence by Google and Anthropic’s enterprise focus. This week, the company first launched agentic coding tool Codex and four days later turbo-charged it with a new model called GPT-5.3 Codex.
While the launch itself puts OpenAI on the radar of a business where Anthropic had a head-start, what was more noteworthy is the fact that the new model release came minutes after Anthropic had dropped its own. The company released its latest and most advanced version of Opus, particularly critical for Claude Code agentic AI assistant for powerful, terminal-based coding.
OpenAI describes the new model as one that transforms Codex from an agent that can write and review code to one that can do nearly anything developers and professionals do on a computer, expanding who can build software and how work gets done.” After testing against performance benchmarks, the company claims can create “highly functional complex games and apps from scratch over the course of days.”
As for Anthropic, the latest Opus 4.6 comes three months after they released the 4.5 version. With this release they aim to broaden its capabilities and appeal to a larger customer base. A crucial addition is called “agent teams” which is a team of AI agents that can split up larger tasks into smaller segmented jobs.
“Instead of one agent working through tasks sequentially, you can split the work across multiple agents — each owning its piece and coordinating directly with the others,” says Anthropic. Scott White, their Product Head claims it is like having a team of humans working for you. Segmenting agents allows them “to coordinate in parallel [and work] faster.”
Quite clearly the race is on and OpenAI appears to be battling on two fronts – with Anthropic on enterprise-level solutions and with Google, whose Gemini hit 750 million weekly users.
Published media reports suggested that both OpenAI and Anthropic had originally decided to release their agentic coding tools at the same time. However, for some reason the latter pushed it up by 15 minutes on the Pacific Time and hit the release button at 0945 hours. Was it a moral win or just a game of one-upmanship? Who cares, as long as we are having fun!
OpenAI claimed that GPT-5.3 Codex was 25% faster than their previous GPT-5.2 version, making it the first that was “instrumental in creating itself.” Does this mean the folks working there actually used early versions of the program to debug itself?
This is what their blog post said: “GPT‑5.3‑Codex is our first model that was instrumental in creating itself. The Codex team used early versions to debug its own training, manage its own deployment, and diagnose test results and evaluations—our team was blown away by how much Codex was able to accelerate its own development.”
Anthropic, which hit the press about an hour before OpenAI did (at least that’s what the timestamps on their blog convey) says “We build Claude with Claude. Our engineers write code with Claude Code every day, and every new model first gets tested on our own work. With Opus 4.6, we’ve found that the model brings more focus to the most challenging parts of a task without being told to, moves quickly through the more straightforward parts, handles ambiguous problems with better judgment, and stays productive over longer sessions.”
The hype and hoopla around AI appears to be a given in these baloney times where bubbles and froth seem to be the main melody, not the surround sound.
Getting beyond the hype, the post on Claude notes that the new version mproves on its predecessor’s coding skills. It plans more carefully, sustains agentic tasks for longer, can operate more reliably in larger codebases, and has better code review and debugging skills to catch its own mistakes. And, in a first for our Opus-class models, Opus 4.6 features a 1M token context window in beta.”
But there’s more. Opus 4.6 also integrates Claude directly into PowerPoint via an accessible side panel. Earlier a user could ask the AI tool to create a PPT, but the file would need to be transferred to the correct format to edit the presentation. Now, the presentation can be crafted within PowerPoint, with direct help from Claude, says White.
Overall, it looks like both OpenAI and Anthropic are seeking to expand their userbase at the enterprise level. From software development, they are hoping that the upgrades would be useful to a broader set of knowledge workers in an enterprise. White claims that Claude is used not just by software developers but also product managers and financial analysts. The race is indeed on!
Of course, OpenAI wasn’t done with its launches on this day
Which brings us to the other launch that OpenAI did in a single day. They came out with OpenAI Frontier as an end-to-end platform for enterprises to build and manage AI agents. Being an open platform, it allows users to manage agents built outside OpenAI too. (You can read the blog here)
“Frontier gives agents the same skills people need to succeed at work: shared context, onboarding, hands-on learning with feedback, and clear permissions and boundaries. That’s how teams move beyond isolated use cases to AI coworkers that work across the business,” says another blog post by the company.
OpenAI says users can use Frontier to program AI agents to connect to external data and applications, albeit with fixed limitations, to carry out tasks outside the platform. It works the same way companies manage human staff with an agent onboarding and feedback loop that assists them in getting better over time.
The company says several enterprises such as Oracle, Uber and HP were customers, though for the moment they could have only limited access, which will grow in the months ahead as they add to the plans and pricings. Of course, it is not just these two AI giants who’re in the business of agent management. Salesforce with AgentForce arrived in 2024 while others like LangChain and CrewAI are hot on their heels with recent funding rounds.
Why the sudden rush towards agentic platforms isn’t surprising
Driving a further context around this sudden spurt of activity around agent management platforms, in December Gartner had released a report where it had claimed such platforms to be the “most valuable real estate in AI” and a chapter of the infrastructure story that would make it irresistible for enterprises.
“These platforms don’t just organize agent deployment; they provide means to unify the tools, libraries, security layers, and dashboards needed to monitor performance, mitigate risk, and scale with confidence,” Gartner had said. And OpenAI had shouted from the rooftops that for the company 2026 was all about enterprise adoption. Just that Anthropic started the race a year ago.
CXOtoday is a premier resource on the world of IT, relevant to key business decision makers. We offer IT perspective & news to the C-suite audience. We also provide business and technology news to those who evaluate, invest, and manage the IT infrastructure of organizations. CXOtoday has a well-networked and strong community that encourages discussions on what’s happening in the world of IT and its impact on businesses.
Copyright © 2025 Trivone. All Rights Reserved.
We use cookies to improve your experience on our site. By using our site, you consent to cookies.
Websites store cookies to enhance functionality and personalise your experience. You can manage your preferences, but blocking some cookies may impact site performance and services.
Essential cookies enable basic functions and are necessary for the proper function of the website.
You can find more information in our Privacy Policy and Privacy Policy.