@Chopsyturvey As the saying goes....
Those who know don't talk, and those who talk don't know.
Don't believe everything coming out of certain countries who rely on stealing technology more than invention, and don't assume that those who don't talk don't have more to say.
@histories_arch Now imagine Canis Pugnax (the Cane Corso and Neapolitan Mastiff ancestor) being 24"-28" tall and weighing around 110 standing beside them and charging the enemy lines. They must have been formidable in their day.
@JeanieRose1960@IRanMediaco I don't make it a habit to reply here, but this is so outrageous that I couldn't let it stand. You sound more like Tokyo Rose than some JeanieRose here. What you are saying is complete fabrication. If you believe what you are writing, you really should get better sources.
@NoSpinZA@IRanMediaco I think the best negotiators we have right now are those B52s flying over Iran. You don't fly those unless your enemy is to a point they can't do anything about it. I really hope for their sakes, and our bank accounts, the Iranians negotiate.
@sama Multi-agent systems are powerful.
But as reasoning spans agents and tools, maintaining intent continuity and preventing drift becomes the hard problem.
In the agentic era, orchestration is a control-plane challenge, not just a capability race.
@SciTechera This is rather misleading.
Sarvam Vision is a 3-billion parameter model. Because it is smaller and hyper-focused on OCR and document parsing, it can be more accurate in those specific tasks than a trillion-parameter "generalist" model like GPT and Gemini.
@Shawn_Farash Really, I think the best way to effect change is to start with yourself. Stop watching all of the games. Stop purchasing any of the merchandise. That will get their attention.
@pankajkumar_dev I suggest both, and add Gemini to the mix for your UI tester. ChatGPT Codex makes a wonderful architect, and Claude Opus makes an amazing engineer. Why not use each to each's strengths?
Here is an article talking about that: https://t.co/eW9A41YWho
@mitchellh I've seen good, and bad, results from them all. I normally have more than one of the CLI tools open at the same time and move back and forth if one gets stuck. Gemini 3 was able to "save" a Claude miss, Codex (ChatGPT) something Gemini missed, etc. I find each has pros and cons.
@GeneralMCNews This is about the same scale as the Manhattan Project. It's that big, and we all need to be pulling in the same direction. For that reason, I get it. It's like national defense folks. Now, I'd have liked a shorter term on it, but I'm glad that at least there is one.
@The_x_Truth@GeneralMCNews This is akin to the Manhattan project. It's important that we all pull together. Given that, I understand consolidating power, for now at least
@RichestManAliv2@bindureddy I've tested both Gemini 2.5 Pro and Claude 3.7 in a hypothetical coding test. I've seen various other tests that do show Gemini is better at coding some things, but for this specific text (Kotlin app on a PC) Claude edged it out. In general non-coding tasks though, Gemini wins.
@tangyoj @LeadingReport I'm in a field where H1B visas play an important role. The truth is, we need some talent above and beyond what we grow domestically. Now, there needs to be good rules in place, but it should not be stopped. Look at all of the foreign talent that helped in AI for example.
@Er0tal @toxicity_dcs @DANNY_THE_NOSS@truthpole BINGO assuming you are talking about the one where Tyson jabbed to Paul's waistline. His waistline folks. You don't fake a jab to someone's waistline!!!
@Ojoyd@eapotterjr@CollinRugg@mytruvote At some point in HISTORY, all ethnicities have been persecuted by others. The African Americans just happen to be more recent, but that's still HISTORY. What that woman is doing is helping no one. When are we going to stop judging each other by the color of our skins?