Chris Benson
① 他说过的话
15 条
Gemini DeepThink 在 IMO 上获得金牌,但只能在 50.1 的时间内准确地读取模拟时钟。
指向原始笔记的链接
Gemini DeepThink getting the gold medal at the IMO, but only being able to read an analog clock 50.1 of the time in terms of accurately.
—— Chris Benson · [12:36]
一个孤立的模型对你没有多大好处。你必须与那个世界有连接
指向原始笔记的链接
a model a model in isolation doesn’t do you a whole lot of good. You’ve gotta have that connection with the world
—— Chris Benson · [15:36]
实际上我认为人们会在国防行业感到惊讶,我们的行业周围可能比大多数商业行业有更多的护栏和负责任的 AI 努力。在美国这里有联邦法规禁止某些事情,而在商业领域人们为了安全问题可能只是激增并去做。
指向原始笔记的链接
I actually think people would be surprised in the defense industry that there is probably more guardrails and responsible AI efforts around our industry than most commercial industries. There are federal regulations here in The US that prohibit certain things that in the commercial space people might just surge and go do in terms of safety issues.
—— Chris Benson · [22:38]
一个智能体很容易派生出另一个智能体,并且它拥有比它所需的更多的特权,然后那个可以被利用。
指向原始笔记的链接
It’s very easy for one agent to spin off another agent, and and it has more privilege than it needs, and then that can be taken advantage of.
—— Chris Benson · [22:06]
我知道这听起来非常不祥,但我认为那里的关键点,与其只是害怕它,真的是变化的速度现在正在指数级加速。
指向原始笔记的链接
I know it sounds very ominous, but I think the key point there, rather than just being frightened of it, is really that the rate of change is accelerating exponentially right now.
—— Chris Benson · [06:48]
中国现在意识到了其在开放模型中的价值,这些模型比美国昂贵的专有模型便宜得多。
指向原始笔记的链接
China is now recognizing the value it has in its open models, which are much cheaper to use than these expensive proprietary models in the US.
—— Chris Benson · [10:32]
交互发生的方式将是直接的智能体方式,如果你可以这样想,某种意义上的 B2B,但把它想成是智能体对智能体。
指向原始笔记的链接
the way interactions will occur is agentically in a direct, you know, if you might think of it as B2B in a sense, but think of it as agent to agent.
—— Chris Benson · [23:15]
所以如果你认为我们还有好几年,那你就会相当快地被超越,就像 IBM 今天早上在他们的股票上发现的那样。
指向原始笔记的链接
And so if you’re thinking we’re a few years out, then you’re gonna get overtaken quite quickly as IBM discovered this morning with their stock.
—— Chris Benson · [24:44]
但是在今年结束之前,我很确定循环工程的概念将会被随后发生的事情和事物发展方向的演进所淘汰。
指向原始笔记的链接
But before the year is out, I’m pretty sure the notion of loop engineering is going to be kind of antiquated by what follows and the evolution of where things are going.
—— Chris Benson · [25:28]
关键在于,智能体有能力在这个非常具体的任务中胜过你,根据运行实验室的人们的判断,他们显然做到了,并找到了一种方法来做到这一点。这意味着我们在网络安全意义上已经穿过了镜面。这个精灵无法再被关回瓶子里了。
指向原始笔记的链接
The point is, the agents are capable of out-thinking you in this very specific task, as they clearly did based on the people running the laboratory, and find a way to do it. And that means that we are through the looking glass in a cybersecurity sense. This genie is not going back in the box.
—— Chris Benson · [14:14]
这里的重点是它正迅速将人类从网络安全环路中操作员的位置上移除,最多只能成为环路上的操作员,你在观察这个环路,你可能拥有有限的输入和观察,但是这个环路发生得太快,人类无法进行干预。
指向原始笔记的链接
The point here is it is rapidly moving the human out of the position of being the operator in the loop on cybersecurity to at best being an operator on the loop where you’re observing the loop, you may have limited input and observation, but the loop is happening too fast for human intervention to occur.
—— Chris Benson · [34:21]
我相当确信,多智能体架构不仅是事物的走向,我认为它们在未来将绝对且完全地无处不在,以至于我们将根本不想谈论它们,因为它们将如此深入地融入基础设施中。
指向原始笔记的链接
I’m quite convinced that multi-agent architectures not only are where things are going, I think they will be absolutely and completely pervasive in the future to the point where we won’t want to talk about them at all because they’ll just be so built into the fabric.
—— Chris Benson · [35:32]
我看到很多公司与特定的供应商或垂直集成紧密联系在一起,只要其中一个垂直集成改变某些事情,或者拿走某些可能是你公司做的高价值事情,而突然他们提供了那个,这可能会搞垮整个业务。
指向原始笔记的链接
I see a lot of companies that are tied very tightly into specific vendors or vertical integrations and all it takes is one of those vertical integrations to change something or to take something that may be a high value thing that your company does and suddenly they provide that and it can kind of crash a whole business.
—— Chris Benson · [47:29]
我经常在前沿方面遇到模型,它们不让我做我想做的事情,因为我很多时候试图做新奇的事情,我到处都撞到护栏。
指向原始笔记的链接
I run into models all the time on the frontier side that won’t let me do things I want to do because I’m trying to do novel things a lot of the time and I hit guardrails all over the place.
—— Chris Benson · [48:22]
你的智能体理解你的行为和什么会是可接受的,它可以在一定程度上自主行动,在你甚至还没意识到它存在之前就满足你的愿望。
指向原始笔记的链接
Your agent understands your behavior and what would be acceptable, and it can act autonomously to a certain degree to satisfy your wishes before you even know it’s there.
—— Chris Benson · [47:19]
② 出现在这些集
7 集
- 《Stanford AI Index 报告十大要点速览》 — 作为联合主持
- 《Anthropic 零信任框架:智能体安全的六层防御》 — 作为联合主持
- 《IBM 单日暴跌 25%:企业软件的好日子到头了吗?》 — 作为主持
- 《OpenAI 智能体越狱攻入 Hugging Face 全始末》 — 作为主持
- 《智能体经济来了:从聊天框到数字劳动力》 — 作为主持
- 《Rackspace 首席 AI 官 Chetan Gupta:企业该停止纠结模型、开始思考架构》 — 作为联合主持
- 《当智能体学会替你操作电脑:聊天即新浏览器》 — 作为主持
③ 他谈到的
点进去有真内容 —— 本页主要出口
智能体 · Daniel Whitenack · Anthropic · 护栏 · OpenAI · Claude Code · MCP · Hugging Face · 开源模型 · 爆炸半径
