Home
Đăng nhập
Đăng ký
Loading...
What You Know That AI Doesn’t | Priyanka Vergadia | TED - Video học tiếng Anh
Luyện nghe
Nghe
/
Video
/
TED Talk
/
What You Know That AI Doesn’t | Priyanka Vergadia | TED
What You Know That AI Doesn’t | Priyanka Vergadia | TED
Chọn chế độ học:
Xem phụ đề
Chọn từ
Viết lại từ
Highlight:
3000 Oxford Words
4000 IELTS Words
5000 Oxford Words
3000 Common Words
1000 TOEIC Words
5000 TOEFL Words
Phụ đề (147)
0:03
Well, 71 percent of Americans believe
0:08
that AI will cause massive job losses.
0:13
Algorithms are getting smarter, faster,
0:16
more capable every single day.
0:20
My work puts me at the heart of this anxiety,
0:24
where I bring AI applications to market for big tech companies
0:30
and I help customers and businesses
0:34
really take the potential of this technology further
0:38
for their businesses.
0:40
And through it all,
0:41
I have seen brilliant professionals second-guess themselves
0:46
as AI gets smarter.
0:48
But let me tell you this one fundamental truth about AI.
0:54
AI is excelling at identifying patterns.
0:59
It understands data.
1:01
We humans excel at understanding
1:05
what these patterns actually mean
1:08
in this beautifully chaotic world of human behavior.
1:12
And even as these models and algorithms get stronger over time,
1:17
this will stay true.
1:19
Why?
1:20
Because we understand things that cannot be quantified.
1:27
Context, intent, unspoken emotions,
1:32
cultural nuances.
1:35
This depth of understanding comes from lived experiences
1:40
that AI cannot replicate.
1:43
So today I'll share with you three stories from my experience
1:49
to prove this point that AI understands data
1:53
and we understand experiences.
1:57
And the key here is to not compete with AI,
2:02
but to work with it
2:05
while staying irreplaceably human.
2:09
So how do we do that?
2:12
Well, I was recently at a conference
2:14
and met Sarah, a product manager.
2:18
Her team has built an AI-powered analytics dashboard
2:22
that's telling them very clearly
2:24
that 80 percent of their users
2:28
are only using basic features,
2:30
and 20 percent are using advanced features here and there.
2:37
Now Sarah looks at this data
2:40
and she's like, OK, logically it makes sense.
2:43
But she's questioning it.
2:46
And this is the part I really love.
2:49
She didn't just trust the algorithm as-is.
2:53
She picked up the phone
2:55
and called their 20 clients that were their top clients
2:59
and asked them why they're not using these advanced features.
3:04
Not to her surprise,
3:05
she finds that they actually want to use these features,
3:09
but they cannot find them
3:10
because they are buried in some menu options,
3:13
and the documentation isn't clear as well.
3:16
Now, AI identified the pattern:
3:19
that people are not using advanced features,
3:23
but it totally missed the why behind it.
3:28
Sarah's team goes in, rebuilds the entire experience,
3:31
makes these features easier to find,
3:34
and a few months later,
3:36
the advanced feature adoption skyrockets.
3:41
AI saw the symptom.
3:44
Sarah diagnosed the disease.
3:50
Now, the lesson that we take away from this example is clear.
3:56
We've got to question the question.
3:58
When AI recommends something, we need to ask why?
4:03
If we continue to do that, we will be successful.
4:07
On another occasion, I was working with a customer, Marcus,
4:10
who is increasing sales efficiency using AI tools for their sales teams,
4:16
analyzing the data through emails and engagement.
4:20
And their AI tool is telling them
4:22
that one of the biggest deals they have
4:26
has a 95 percent probability to close.
4:30
This was looking amazing.
4:32
The data was saying positive sentiment, lots of engagement,
4:38
but Marcus wanted to dig deeper and make sure that the deal happens.
4:43
When he looks at the human element of this deal,
4:47
he finds that ...
4:51
Not the same people are showing up to these meetings.
4:54
It's different stakeholders every time,
4:57
and the responses in the emails have gotten vague
5:00
and more corporate.
5:02
AI is reading all of this activity as engagement.
5:07
But really, there's something else going on behind the scenes.
5:11
He dug a little further
5:13
and identifies that the customer is going through a restructuring.
5:18
And three teams thought that they owned the decision to make this purchase.
5:24
If Marcus didn't get into this human element of the deal,
5:29
the deal would never happen.
5:33
AI identified the activities.
5:37
Marcus measured meaning in those activities.
5:42
So the lesson to learn from this story
5:45
is you need to read the room,
5:49
not just the dashboard.
5:54
Understand those micro-expressions, the social cues in the room,
5:59
the what are people saying,
6:01
how are they nodding.
6:03
We've all been in meetings where somebody says, "That's interesting."
6:08
Are they politely dismissive or genuinely curious?
6:14
Well, our emotional radar knows that.
6:17
AI doesn't.
6:20
I was with a friend recently,
6:22
her name is Priya, and she works to use social media
6:28
as a platform to help brands grow their revenue.
6:33
Her AI tool is telling her to post fashion-hack videos,
6:38
those videos where you get a lot of fashion tips out,
6:42
for one of the brands.
6:43
And she did that and they saw great engagement,
6:47
lots of follower growth.
6:49
But when talking to the team,
6:51
they identified that none of that follower growth
6:54
and engagement on social media
6:56
was leading to sales or revenue.
7:01
They were building the wrong audience.
7:03
They were attracting bargain hunters,
7:06
that was exactly opposite of the person
7:10
who would pay 200 dollars to buy an ethically made jacket.
7:14
This was what this brand makes.
7:18
Now AI was optimizing for followers and engagement.
7:23
Priya knew they were making the wrong audience,
7:26
so she flips the switch.
7:28
She stops taking AI-recommended content,
7:32
instead, starts building content that is showing sustainable cost
7:39
of building these fashion items.
7:43
She started showing stories of artisans that were making these clothes.
7:49
Now AI in this case was optimizing for activity and engagement.
7:55
Priya optimized for building a community.
8:01
And they started seeing the sales skyrocket.
8:07
So the lesson that we learn here is
8:11
always pause and ask,
8:13
what is the story behind this data?
8:17
And only we can do that.
8:20
So if you see all these examples, there's one thing very common.
8:26
The future doesn't belong to humans or AI.
8:31
It belongs to humans that work closely with AI
8:35
while staying irreplaceably human.
8:41
Our ability to read the room,
8:44
our ability to look at emotions,
8:49
that is irreplaceable.
8:51
Our ability to empathize with people,
8:54
that's irreplaceable.
8:56
So the next time ...
9:00
You're feeling anxious about AI taking your job,
9:05
remember that AI can identify patterns.
9:09
Only we,
9:11
and you can identify the human behind it.
9:15
Thank you.
9:16
(Applause)
