So, I'm not happy. Google has done another update, and I love updates, I love new features, and things getting better. They released Gemini 4 Argon; which is new and smarter and ... well, you can see what they said below.
See, up until that release (and I'm going to attribute the change to this update, as that's when what I noticed stopped working), I had been using Spark (now called 'tasks' so it's harder to navigate) to build up a system for me where it takes information I have from a tab on a Sheet and then sends an email to specified people, then waits for their replies and updates another tab with their responses accordingly.
It had been going for a month, and then since Sunday I started having errors. I noticed that replies weren't entered, and then I'd try to have a conversation with Gemini about why it hadn't and it just said "yup, i see those answers, and yup there should be the following data in the following cells" - but when I asked it to do it, it said it would, but then nothing changed. I asked it what's wrong and it would say that there was a permission denied.
Apparently I could make the sheet 'anyone with the link' but I have private information on it I don't want ANYBODY to have, so I kept it locked down.
I've tried the 'appscript' option to see if it'll work but haven't had a chance yet to see if it triggers automatically.
C'mon Google, I like updates and all, but TELL US when something won't work anymore.
You can see in the 'connections' for Workplace that Sheets is no longer listed. It can read data from Sheets but just can't edit it anymore.
So annoying to have built this system that works for over a month and start to trust it's reliability only to have it yanked away.
Anyways, I'm sure 4 does some other awesome stuff (just don't ask it to edit a Sheet).
Today, Google rolled out Gemini 4 Argon globally for developers and enterprises via the paid tier of the Gemini API, for Google AI Ultra subscribers in Google Antigravity and Google AI Studio, and for Google AI Ultra and Pro subscribers in the Gemini app and AI Mode in Search. Argon excels at deep work in complex, real-world tasks in many domains including coding, finance, legal, cybersecurity, and science. Its depth and precision on multi-step tasks leads Argon to outperform other frontier models on benchmarks across domains.
Teams at Google, enterprise customers, and in the research community have already been using Argon to land significant impact:
- Solving complex engineering tasks: Over the past few weeks, software engineers have been shipping high-quality code faster than ever. Argon in Antigravity has helped teams increase agentic code submissions by more than 35% while reducing code rollbacks by nearly a third. For example, Argon helped the Chrome team separate core webpage handling from a ~6 million line legacy folder of browser features held within Chromium’s 40 million-line codebase, completing months of work in weeks.
- Transforming critical business and professional workflows: Enterprise customers and startups in our early access program (including Deloitte, LexisNexis, Ambience Healthcare, and Isomorphic Labs) are already seeing strong wins with Argon for use cases like deep research, document synthesis, longitudinal data analysis, and large-scale codebase reasoning.
- Strengthening cybersecurity defense: Starting Tuesday, Google is expanding its Fairwind Program to hundreds of additional partners, building on early testing where Wiz used Argon through its "Scan for Good" initiative to chain complex bugs and patch critical flaws protecting hundreds of thousands of private patient records.
- Accelerating scientific breakthroughs: Argon is enabling researchers to make new discoveries across domains — from nearly doubling quantum error-correction throughput on AlphaQubit to discovering new mathematical solutions for clean fusion with a professor at the University of Maryland, and more.
Source: https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-4-argon/
Comments
Post a Comment