Tag: coding

Weekend roundup; expanded Sora, security research and the battle for India

For a limited time, Sora is available without invite codes for select countries, but the 30 generations per day limit may have to go.
No more invite codes for select countries in Sora 2, and bevy of new features. (Picture: generated)

Sora 2 expands, is now available without invite codes
Following the massive success of the Sora 2 video generator, OpenAI is opening up the service for those without invite codes in the USA, Canada, Japan and Korea «for a limited time.» Simultaneously, they are announcing reusable characters that can feature in more than one video and an easier way to stitch videos together. If that wasn’t enough, OpenAI is adding more video generations for power users hitting the 30-per-day generation limits and letting them pay for more gens. They are also musing about letting rightsholders get compensation for the reuse of their characters, as a means of getting paid for your work on the platform. They do warn that 30 gens needs too many GPUs and will be throttled at some stage.
More at: MacRumors and a Twitter announcement, list of available countries.

OpenAI reveals security research agent in beta
The new agent, Aardvark, will look through code repositories at scale almost like a human would, and find errors and exploits before the bad guys do. It will continually analyze your source code and find vulnerabilities. The agent has already been used to find «numerous» vulnerabilities in open source software, and OpenAI will provide pro bono scanning to «select, non-commercial» OSS systems. Aardvark is not being widely released, existing instead as a private beta inside OpenAI’s offices, kind of like Google’s CodeMender.
More at OpenAI’s announcement and ZDNet.

Read on for more!

Continue reading “Weekend roundup; expanded Sora, security research and the battle for India”

OpenAI’s 2025 Dev Day with Altman livestream incoming

Speculation is rife as to what Altman might announce at the livestream.
Will it be a browser, a new image model, or the highly anticipated AI device? It’s too soon to tell. (Picture: generated)
The October 6. stream will be an excellent moment to announce product news for the OpenAI CEO, but only a few items remain on their to-do-list.

Nothing has been announced but a cryptic tweet promising «new ships,» which could mean anything from new models to new modalities:

Continue reading “OpenAI’s 2025 Dev Day with Altman livestream incoming”

OpenAI tops ICPC coding contest for students, Google finished second

OpenAI solved 12 of 12 problems with vanilla GPT-5. Google had a custom model and solved 10.
OpenAI says they will now focus on scientific discovery. (Picture: OpenAI)
ChatGPT solved all 12 of 12 problems in the 2025 International Collegiate Programming Contest (ICPC) — an algorithmic programming contest for university students.

That result would have given it first place if it were human, as the best college teams only solved eleven.

Google also participated with a custom Gemini 2.5 Deep Think and earned Gold status, solving 10 of the problems and finishing second, Google claims.

Continue reading “OpenAI tops ICPC coding contest for students, Google finished second”

OpenAI announces GPT-5 Codex

GPT-5 Codex is slightly better than vanilla GPT-5 in benchmarks.
OpenAI is especially proud of the code review function in the new Codex. (Picture: OpenAI)
Savvy users have been using GPT-5-high with the Codex CLI (Command Line Interface) on their terminals for weeks, and consensus seems to be that it competes well with Claude.

Now, OpenAI is launching a custom, optimized version of GPT-5 for the Codex coding agent that they say is faster, more reliable and more steerable than before.

Continue reading “OpenAI announces GPT-5 Codex”

Google hires top execs, team from Windsurf — upending OpenAIs deal

Google hires top execs and talent from Windsurf
Just as talks with OpenAI ended, Windsurf turned to Google. (Picture: Windsurf)
OpenAI had been negotiating a $3 billion to acquire the agentic coding platform, but Google just snagged their top executives to work in their field for its Gemini platform.

The deal will see Windsurf CEO Varun Mohan, co-founder Douglas Chen and a small team join Google’s DeepMind division.

Also licensing key tech
Further, Google will invest $2.4 billion in a non-exclusive deal to license Windsurf’s technology, reports Reuters, among others.

OpenAI had been in long-winding talks to buy the company, in its biggest deal yet, and many said it was just around the corner, as late as in May, 2025.

Continue reading “Google hires top execs, team from Windsurf — upending OpenAIs deal”

OpenAI’s Codex now available to ChatGPT Plus users

ChatGPT Plus-tier gets access to Codex!
Wider availability for Codex likely means even more pressure on the coding market. (Picture: Chatgpt.com)
Caught this morning, there seems to be a new option in the sidebar at Chatgpt.com for the new Codex coding model — meaning it has expanded access.

Codex is the latest coding agent from OpenAI that runs on a modified o3-model.

Super-coding agent
It can generate several instances of code from your prompts, and even run them in a sandbox to select the best/most efficient version.

OpenAI says it can complete tasks autonomously that would otherwise take hours or days to finish, and they are using it themselves to offload repetitive tasks.

The Plus membership for ChatGPT is $20 a month, and Codex launched as a «research preview» in May for Pro users, who fork out $200 a month.

Update: It appears Codex now also has Internet acccess, which is off by default and comes with a stern warning.

See also: teknotum on the Codex launch, and the announcement thread on X.

Anthropic claims world’s best coding AI with Claude 4 Opus and Sonnet

World's best coding model? According to Anthropic, yes, of course.
Anthopic’s new agentic, thinking and reasoning models are great for coding, and plays Pokemon for 24 hour runs. (Picture: Anthropic)
Opus 4 can sustain almost a full work day of focused coding work, while Sonnet 4 is supposed to be excellent for thinking and reasoning.

Both models produce near-instant responses to queries, but can turn to reasoning and thinking for more demanding requests.

World’s best on coding?
Anthropic claims Opus is «the world’s best coding model,» and it edges out Gemini 2.5 Pro, o3 and GTP 4.1 on SWE-bench Verified, but cannot surpass OpenAI’s o3 on certain PhD-level benchmarks, according to TechCrunch.

Continue reading “Anthropic claims world’s best coding AI with Claude 4 Opus and Sonnet”

OpenAI debuts Codex, an AI coding agent, further disrupting the software industry

Codex can run your code multiple times until it finds the best version.
Codex can run your code multiple times until til finds the best version. (Picture: OpenAI)
In a world where 30% of Big Tech code is written by AI already, OpenAI says it wants to join the club for «virtual teammates,» and it’s getting ever closer with today’s launch.

OpenAI’s Agents Research Lead, Josh Tobin tells TechCrunch that the company wants software agents to complete tasks autonomously that would otherwise take hours or days to finish, in anything from one to thirty minutes, and OpenAI is already using the model to offload repetitive tasks.

Continue reading “OpenAI debuts Codex, an AI coding agent, further disrupting the software industry”

Big Tech embraces AI coding, hitting 30% of software

Major big tech businesses are doing substantial amounts of AI coding.
Gemini 2.5 Pro opens even more avenues for coding, says Alphabet. They are taking that to heart. (Picture: Google)
With Satya Nadella’s announcement overnight that Microsoft uses AI to code around 30% of their software, AI coding has come of age. Other Big Tech companies have also reported similar numbers lately.

— I’d say maybe 20%, 30% of the code that is inside of our repos today and some of our projects are probably all written by software, said Microsoft CEO Satya Nadella during a conversation at LlamaCon, Meta’s AI developer conference, according to CNBC

Microsoft says they have made progress with Python, and less with C++ projects, but he is still pointing that some projects could be entirely written by AI.

Continue reading “Big Tech embraces AI coding, hitting 30% of software”

Anthropic: Virtual employees will arrive next year

Virtual employees are step uup mere agents, say Anthropic.
Virtual employees are a step up mere agents and could be roaming the offces pretty soon. (Picture: Anthropic)

In a recent interview with Axios focusing mostly on security issues, Anthropic said «virtual employees» will be a step up from using mere «agents» on corporate networks.

This will be the next AI innovation, said Jason Clinton, the company’s chief of information security.

Whereas agents can focus on specific, programmable tasks, acts with some autonomy and of course require oversight, a «virtual employee» takes it a step further, with having their own memories and their own corporate accounts and passwords.

This is a major headache for cybersecurity, Clinton further explained, about oversight and hackability of the new employees.

Go read the full story here – at Axios, and check out Anthropics research on agents.