Alessandro Perilli

22.1K posts

Alessandro Perilli banner
Alessandro Perilli

Alessandro Perilli

@perilli

Vice President, Research @ IDC. I lead the AI Strategies research team. We focus on emerging AI tech, global AI market trends, and the state of enterprise AI.

London Katılım Ekim 2010
4 Takip Edilen8.5K Takipçiler
Sabitlenmiş Tweet
Alessandro Perilli
Alessandro Perilli@perilli·
Lots of people are asking what I do at IDC, so let me tell you the plan. The research program I lead focuses on two opposite ends of a spectrum. On one side: the newest, most innovative ideas in AI. I won’t look only at the technology layer, but also at how that layer affects business and organizational models, professions, welfare strategies, the economy, and more. Think of this as an incubator program for bold new concepts. We want to capture what’s next in AI before it becomes obvious to everyone. If and when these ideas become mature enough, we’ll create new vertical research programs to fully explore how the market absorbs the technologies and the models behind them. On the other side of the spectrum: how large end-user organizations adopt AI, starting with CIOs, CISOs, CEOs, and boards. I will look at how they make decisions, how they perceive the AI opportunity and the market landscape, who is doing something differently, and how those decisions make them successful. I will also look at how they rethink the role of their organizations to address the challenges AI creates. Think of this as an attempt to capture the playbook of AI winners, rather than focusing only on the challenges posed by new, disruptive technologies. To move between these two extremes, my team and I will use the invaluable data IDC allows us to collect at a global scale. We can offer a macro view of who is doing what, where, and at what scale. And that view is not limited to North America. My team is in Europe, the Middle East, India, Singapore, and South Korea. What I described is a very ambitious, monumental mission. Obviously, I can’t do this alone. Beyond my team, I can count on an amazing group of IDC analysts all around the world. I’ll work with all of them to understand how the world is changing and what our clients must do to win in the age of AI. Isn’t this the best job in the world?
English
0
1
4
739
Brett Winton
Brett Winton@wintonARK·
@perilli 99.9% of meetings will be between AI agents Those sending underpowered agents will see marginally diminished power/prestige/access
English
1
1
1
765
Alessandro Perilli
Alessandro Perilli@perilli·
Phase 1: Every participant will show up to a meeting with their own AI agent. Phase 2: When participants cannot attend, they will send their AI agents in their place. Phase 3:
English
1
1
1
778
Alessandro Perilli
Alessandro Perilli@perilli·
I need some help answering a burning question (not official IDC research, but I promise I'm going somewhere). I hope both technical and non-technical people will answer. Here’s the question: Let’s say you have two AI agents available on your computer and phone. They are almost identical: - They have identical budget limits (for example, you can spend $3K per month on each). - They have identical capabilities and access to tools ( browsers, email, calendars, design apps, coding apps, everything). The only difference is this: - Agent 1 has the voice of Paul Bettany as J.A.R.V.I.S. in the Iron Man movies. - Agent 2 has the voice of Scarlett Johansson as Samantha in Her. Imagine yourself using both throughout the day, and tell me: Do you use them for the same tasks and projects? Do you give them the same commands? (resharing for broader reach is welcome, thank you!)
Alessandro Perilli@perilli

You have 2 AI agents available on your computer/phone. They are identical, but: - Agent 1 has the voice of Paul Bettany as J.A.R.V.I.S. in the Iron Man movies. - Agent 2 has the voice of Scarlett Johansson as Samantha in Her. Do you use them for the same tasks and projects?

English
1
1
1
322
Alessandro Perilli
Alessandro Perilli@perilli·
You have 2 AI agents available on your computer/phone. They are identical, but: - Agent 1 has the voice of Paul Bettany as J.A.R.V.I.S. in the Iron Man movies. - Agent 2 has the voice of Scarlett Johansson as Samantha in Her. Do you use them for the same tasks and projects?
English
1
0
0
606
dax
dax@thdxr·
someone make this for me it's just a mic
dax tweet media
English
113
5
998
142.9K
Alessandro Perilli
Alessandro Perilli@perilli·
Alessandro Perilli@perilli

The UI approach for many agentic AI is absolutely chaotic right now. We need a minute to rationalize the access and organization of the many exceptional features that have been released so far. This is what I would like to see ASAP: 1. The same identical experience across desktop, mobile, and web. Every project, every task, every chat, every site, every artifact. Right now, I have to waste 5 minutes finding a thread/task because I can't remember if I created it on the web or the desktop app. Both have Projects, but they contain different things. I cannot move a thread from the assistant to the agent or vice versa. And so on and on and on. 2. The capability to remotely control agents via web, not just via mobile. I understand the complexities of implementing multi-account environments in desktop apps, but I don’t understand why I need to use my phone to control an agent that isn't on the machine I am using. I should have to resort to a tunneled remote desktop to do this. 3. No more distinction between chat, work, and code. Code is work. A chat can turn into work. Every agent is capable of just chatting, creating a presentation, and writing code. Nobody will eventually use just AI assistants anymore. This is an internal branding problem that is seriously hurting user experience. These agentic AI platforms are this close to being exceptional. Let's not make a mountain out of a molehill. And let's not lose the long-term focus on the user experience.

English
0
0
0
42
Tibo
Tibo@thsottiaux·
How can we make ChatGPT better. It’s already on your phone, but what do you wish would happen more or less of when you open the app. How can it bring you more joy?
English
4.6K
122
5.1K
772.9K
Alessandro Perilli
Alessandro Perilli@perilli·
@simonw Maybe we'll return to to tiny circles of friends, like when I was a kid. 5-6 people you have known for a while that occasionally meet to talk about whatever is new. Only, done in public for everyone else to observe.
English
0
0
0
81
Simon Willison
Simon Willison@simonw·
49 replies to this and hard to be confident that ANY of them were posted by an actual human
Simon Willison@simonw

I am begging @AnthropicAI please add an automated test suite somewhere that ensures Claude Code on the web never blocks me from cloning or interacting with other public repos from within my existing sessions

English
48
3
198
41.5K
Alessandro Perilli
Alessandro Perilli@perilli·
@athyuttamre When the time comes, I'll want the capability to define this custom instructions: x.com/perilli/status…
Alessandro Perilli@perilli

Agentic AI “computer use” is not a gimmick. If you look a little further ahead, you can see it as a critical building block for voice UIs. First, if you haven’t tried the new instant, bidirectional live voice available in some AI systems, please try it now, or this post won’t make sense. The experience is nothing like voice dictation or previous, turn-based voice interactions. It takes a little time to get used to having an actual conversation with your AI, especially because you still try to respect the turns, you still try to remain as silent as possible between turns, and you still try to think hard about what to say before speaking. All that takes away from the spontaneousness of human communication. But none of that is necessary anymore. You can freely ramble in the same way you would with another person. And the more you try it, the more you realise that you are experiencing something truly big, truly transformative. Now, assume we’ll have that for every major AI agent. Second, assume that “computer use” will become ten times faster than it is today. The AI agent will open apps, click elements, and move windows at a speed far superior to that of a person. We have already seen significant progress in less than a year. Finally, connect the dots by assuming that the conversation you are having with your AI will lead to proactive behaviour: You are chatting with your agent and ask for the news of the day, the performance of a stock, the chart your CFO sent you, or an email whose exact wording you want to check. The agent answers by voice, yes, but also proactively launches the relevant apps and displays those digital artefacts for you. The screen populates, in fractions of a second, with all the information you need to see at that moment, then clears again as you move on to something else. If you can imagine that, you can see why progress in “computer use” is so important to watch. And if you are a software vendor willing to bet that we are exceptionally close to this reality, then you probably want to ask yourself: Can my software be controlled in this way, or does its current UI impede that flow?

English
0
0
1
84
Atty Eleti
Atty Eleti@athyuttamre·
If ChatGPT had voice-specific custom instructions, how would you use it?
English
112
8
267
14.7K
Alessandro Perilli
Alessandro Perilli@perilli·
Agentic AI “computer use” is not a gimmick. If you look a little further ahead, you can see it as a critical building block for voice UIs. First, if you haven’t tried the new instant, bidirectional live voice available in some AI systems, please try it now, or this post won’t make sense. The experience is nothing like voice dictation or previous, turn-based voice interactions. It takes a little time to get used to having an actual conversation with your AI, especially because you still try to respect the turns, you still try to remain as silent as possible between turns, and you still try to think hard about what to say before speaking. All that takes away from the spontaneousness of human communication. But none of that is necessary anymore. You can freely ramble in the same way you would with another person. And the more you try it, the more you realise that you are experiencing something truly big, truly transformative. Now, assume we’ll have that for every major AI agent. Second, assume that “computer use” will become ten times faster than it is today. The AI agent will open apps, click elements, and move windows at a speed far superior to that of a person. We have already seen significant progress in less than a year. Finally, connect the dots by assuming that the conversation you are having with your AI will lead to proactive behaviour: You are chatting with your agent and ask for the news of the day, the performance of a stock, the chart your CFO sent you, or an email whose exact wording you want to check. The agent answers by voice, yes, but also proactively launches the relevant apps and displays those digital artefacts for you. The screen populates, in fractions of a second, with all the information you need to see at that moment, then clears again as you move on to something else. If you can imagine that, you can see why progress in “computer use” is so important to watch. And if you are a software vendor willing to bet that we are exceptionally close to this reality, then you probably want to ask yourself: Can my software be controlled in this way, or does its current UI impede that flow?
English
1
0
4
343
Alessandro Perilli
Alessandro Perilli@perilli·
Agentic AI “computer use” is not a gimmick. If you look a little further ahead, you can see it as a critical building block for voice UIs. First, if you haven’t tried the new instant, bidirectional live voice available in some AI systems, please try it now, or this post won’t make sense. The experience is nothing like voice dictation or previous, turn-based voice interactions. It takes a little time to get used to having an actual conversation with your AI, especially because you still try to respect the turns, you still try to remain as silent as possible between turns, and you still try to think hard about what to say before speaking. All that takes away from the spontaneousness of human communication. But none of that is necessary anymore. You can freely ramble in the same way you would with another person. And the more you try it, the more you realise that you are experiencing something truly big, truly transformative. Now, assume we’ll have that for every major AI agent. Second, assume that “computer use” will become ten times faster than it is today. The AI agent will open apps, click elements, and move windows at a speed far superior to that of a person. We have already seen significant progress in less than a year. Finally, connect the dots by assuming that the conversation you are having with your AI will lead to proactive behaviour: You are chatting with your agent and ask for the news of the day, the performance of a stock, the chart your CFO sent you, or an email whose exact wording you want to check. The agent answers by voice, yes, but also proactively launches the relevant apps and displays those digital artefacts for you. The screen populates, in fractions of a second, with all the information you need to see at that moment, then clears again as you move on to something else. If you can imagine that, you can see why progress in “computer use” is so important to watch. And if you are a software vendor willing to bet that we are exceptionally close to this reality, then you probably want to ask yourself: Can my software be controlled in this way, or does its current UI impede that flow?
English
1
0
0
160
Alessandro Perilli
Alessandro Perilli@perilli·
The UI approach for many agentic AI is absolutely chaotic right now. We need a minute to rationalize the access and organization of the many exceptional features that have been released so far. This is what I would like to see ASAP: 1. The same identical experience across desktop, mobile, and web. Every project, every task, every chat, every site, every artifact. Right now, I have to waste 5 minutes finding a thread/task because I can't remember if I created it on the web or the desktop app. Both have Projects, but they contain different things. I cannot move a thread from the assistant to the agent or vice versa. And so on and on and on. 2. The capability to remotely control agents via web, not just via mobile. I understand the complexities of implementing multi-account environments in desktop apps, but I don’t understand why I need to use my phone to control an agent that isn't on the machine I am using. I should have to resort to a tunneled remote desktop to do this. 3. No more distinction between chat, work, and code. Code is work. A chat can turn into work. Every agent is capable of just chatting, creating a presentation, and writing code. Nobody will eventually use just AI assistants anymore. This is an internal branding problem that is seriously hurting user experience. These agentic AI platforms are this close to being exceptional. Let's not make a mountain out of a molehill. And let's not lose the long-term focus on the user experience.
English
0
0
1
202
Alessandro Perilli
Alessandro Perilli@perilli·
These are literary forms, and can be ordered by scope and length: Haiku → Sonnet → Fable → Tale → Essay → Novella → Novel → Saga → Myth → Opus The top two positions are open to interpretation and could be swapped. Also, it's not guaranteed that Anthropic is strictly following this logic.
English
1
0
23
1.5K
Simon Willison
Simon Willison@simonw·
I've seen a few people predicting that Opus 5 will be out soon and will be better than Fable 5, but have Anthropic clarified how their relative naming scheme works yet? I assumed it was Haiku < Sonnet < Opus < Fable < Mythos - but is Fable meant to go between Sonnet and Opus?
English
122
4
654
129.6K
Alessandro Perilli
Alessandro Perilli@perilli·
Here's an idea of what the next consumer device could be: I’ve tracked the evolution of synthetic voice technology since I was a teenager, when we were still mainly referring to it as text-to-speech. I never thought we would reach the level of realism we have seen this week. These AI models can now interrupt us almost like humans. To be perfect, they still lack the ability to see the physical cues of the person they are talking to. If voice is going to become the main interface for agentic AI, these models will need to see us. I imagine a small device sitting on a table, equipped with a depth camera, LiDAR, structured light, or perhaps an adaptation of the Wi-Fi CSI sensing technology we’ve seen emerge in cybersecurity research in recent years. Something capable of reading our faces, posture, and hand movements as we speak to our AIs and other people. Imagine the timeliness of the interjections.
OpenAI@OpenAI

Introducing GPT-Live, a new generation of voice models for natural human-AI interaction. Rolling out in ChatGPT starting today. You’ll want to turn the sound on for this one.

English
0
0
2
502
Alessandro Perilli
Alessandro Perilli@perilli·
This is our latest forecast about active enterprise AI agents in the next 3.5 years*. The conversations I had in the last four months at IDC suggest that, collectively, we have not yet internalized the implications of agentic AI at scale. This is why I almost always start my questions by mentioning this forecast. Even when we talk about agents today, in many cases, we are still essentially talking about AI-assisted workflows. It is a lot of human-clicking under the banner of human supervision. But human supervision is a bottleneck, and one with an expensive price tag. I think it will become more evident in the coming months. I am looking forward to more conversations with technology providers building an unsupervised future, no matter how unstable that may be today. *The details abour this forecast is in our new Worldwise Agentic AI Adoption Guide 2026 - my.idc.com/getdoc.jsp?con…
Alessandro Perilli tweet media
English
1
1
1
133
Alessandro Perilli
Alessandro Perilli@perilli·
If the J-space is confirmed, verified across models, and exposed to the outside world, it would open an entire new class of observability, governance, ans security solutions 👀
Anthropic@AnthropicAI

New Anthropic research: A global workspace in language models. Of everything happening in your brain right now, only a tiny fraction is consciously accessible—thoughts you can describe, hold in mind, and reason with. We found a strikingly similar divide inside Claude.

English
0
0
0
191