February 2025 Educational Film // Financial Observations about Artificial Intelligence
An overview of the main players in the field today.
At a time when the upheavals in the world of Artificial Intelligence are overturning our notion of time, Professor Jacquet provides an overview of the main players and their movements.
The world is no longer Gaussian, but extreme, and asset management based on portfolio diversification has had a few wrinkles. In a tech world, the risk-capital approach seems more appropriate: don’t miss out on the next Nvidia!
WEBVTT
1
00:00:07.585 --> 00:00:08.255
Hello on.
2
00:00:08.255 --> 00:00:12.015
Welcome to this educational field whose objective is
3
00:00:12.015 --> 00:00:14.215
to provide you with information
4
00:00:14.315 --> 00:00:17.775
and financial insight into an economic sector,
5
00:00:17.775 --> 00:00:21.975
which is definitely at the forefront of business news.
6
00:00:23.015 --> 00:00:24.215
Artificial intelligence,
7
00:00:25.195 --> 00:00:28.815
the man you see in the picture is named Jensen Huang.
8
00:00:29.435 --> 00:00:32.415
He is the founder of a company which is now very well known
9
00:00:32.415 --> 00:00:33.575
whose name is Nvidia.
10
00:00:33.755 --> 00:00:38.215
Nvidia was created by him in 1993.
11
00:00:38.835 --> 00:00:40.215
He is 30 years old.
12
00:00:40.445 --> 00:00:42.095
He's an experienced engineer,
13
00:00:42.155 --> 00:00:45.175
is very knowledgeable about all these technologies
14
00:00:45.395 --> 00:00:47.775
around microchips today.
15
00:00:48.135 --> 00:00:50.175
I mean, before January 27,
16
00:00:50.795 --> 00:00:54.535
Nvidia was the largest market capitalization in the world
17
00:00:55.005 --> 00:00:59.575
with about $3,600 billion
18
00:00:59.755 --> 00:01:02.015
of market value on a nasdaq.
19
00:01:02.835 --> 00:01:06.055
Now after some equity issues, of course Mr.
20
00:01:06.305 --> 00:01:08.495
Huang has been a little bit diluted
21
00:01:08.795 --> 00:01:12.215
and it seems that he holds 3.6% of his shares,
22
00:01:12.355 --> 00:01:15.055
but 3.6% of this amount money
23
00:01:15.575 --> 00:01:20.055
represents about $130 billion and positions.
24
00:01:20.115 --> 00:01:23.695
Mr. Huang at the number 10 of the Forbes list
25
00:01:23.755 --> 00:01:26.735
of billionaires, which is not bad news for him.
26
00:01:26.835 --> 00:01:28.935
Now, when you observe the evolution of Apple
27
00:01:29.195 --> 00:01:32.855
and Microsoft Giants in this business, in this industry
28
00:01:33.735 --> 00:01:36.495
compared with Nvidia over the last four
29
00:01:36.495 --> 00:01:38.535
or five years, what are the conclusions?
30
00:01:39.105 --> 00:01:42.695
Apple is up by 240%, which is great.
31
00:01:43.815 --> 00:01:46.975
Microsoft is up by 180%, which is great.
32
00:01:47.355 --> 00:01:52.095
Nvidia is up by 2700%, which is outstanding
33
00:01:52.875 --> 00:01:54.975
and rarely observed the stock market.
34
00:01:55.475 --> 00:01:58.055
Now obviously Nvidia is not alone the
35
00:01:58.055 --> 00:01:59.495
planet in this industry.
36
00:01:59.915 --> 00:02:02.975
You have the gamma former G,
37
00:02:03.675 --> 00:02:07.455
but you also have Chad D, pity, open ai, xai
38
00:02:07.635 --> 00:02:11.735
and Tropic menstrual AI and so on and so forth.
39
00:02:12.155 --> 00:02:13.695
My objective in this field is not
40
00:02:13.695 --> 00:02:15.935
to describe the entire ecosystem,
41
00:02:15.985 --> 00:02:20.175
which would be a little bit too long and certainly useless.
42
00:02:20.855 --> 00:02:23.215
I would like to select a few key players first,
43
00:02:23.275 --> 00:02:26.735
the largest ones, the ones which are described already
44
00:02:26.735 --> 00:02:29.775
already, apple, Microsoft and so on and so forth,
45
00:02:30.275 --> 00:02:33.615
but also a few additional key players in industry.
46
00:02:33.905 --> 00:02:35.215
These companies are going
47
00:02:35.215 --> 00:02:39.575
to definitely play a role in the future including open AI
48
00:02:39.595 --> 00:02:41.255
for Chad, GPT, but some others
49
00:02:41.555 --> 00:02:43.935
and now will take the financial perspective.
50
00:02:44.155 --> 00:02:47.695
Now observing the financial metrics of all these companies
51
00:02:48.285 --> 00:02:50.735
will lead us to a conclusion we have
52
00:02:50.735 --> 00:02:54.495
to go a little bit in theory of corporate finance.
53
00:02:55.085 --> 00:02:57.095
It's not going to be a theoretical firm,
54
00:02:57.555 --> 00:03:00.485
but you will understand that a a little bit of refurbishment
55
00:03:00.545 --> 00:03:03.565
of the portfolio theory is probably a bit useful.
56
00:03:04.315 --> 00:03:06.805
It's not going to be a debate on the threats
57
00:03:06.945 --> 00:03:10.565
and ethical issues which are created by the
58
00:03:11.085 --> 00:03:13.445
dramatic evolution of artificial intelligence.
59
00:03:13.705 --> 00:03:15.885
For a few reasons. First, it would take too long.
60
00:03:16.185 --> 00:03:18.085
Second, it's not my purpose.
61
00:03:18.465 --> 00:03:21.805
I'm not legitimate to deep dive into the insight
62
00:03:21.865 --> 00:03:23.325
of all these issues.
63
00:03:23.425 --> 00:03:27.005
So we are going to concentrate on financial metrics.
64
00:03:27.585 --> 00:03:30.525
Now, a few words of course about Chad gt.
65
00:03:30.915 --> 00:03:35.085
Chad GT number three is offered to people in 2022,
66
00:03:35.925 --> 00:03:39.885
a year later end of 2023 chat, GPT-4,
67
00:03:40.115 --> 00:03:42.125
durable plus and so on so forth,
68
00:03:42.745 --> 00:03:44.285
and that's going to be amazing.
69
00:03:45.145 --> 00:03:48.485
The propagation is going to be absolutely outstanding.
70
00:03:48.945 --> 00:03:51.645
How much time does it take to take the first
71
00:03:52.465 --> 00:03:55.925
1 million users Instagram, two and a half months?
72
00:03:56.665 --> 00:04:01.365
Wow, Netflix, three and a half years. Cha five days.
73
00:04:02.265 --> 00:04:05.885
The first 100 million users Instagram two
74
00:04:05.885 --> 00:04:10.645
and a half years Netflix, 10 euro cha, two months.
75
00:04:10.945 --> 00:04:14.325
So it's absolutely unique in the history of high tech.
76
00:04:14.865 --> 00:04:18.885
The speed of propagation of this fantastic application,
77
00:04:18.885 --> 00:04:19.925
which is T.
78
00:04:20.545 --> 00:04:23.365
Now about market values on the NASDAQ
79
00:04:23.785 --> 00:04:25.685
of all these AI firms,
80
00:04:25.825 --> 00:04:28.805
you can take the six largest capitalization.
81
00:04:29.035 --> 00:04:30.965
They are all involved more
82
00:04:30.965 --> 00:04:33.605
or less in the artificial intelligence business.
83
00:04:34.425 --> 00:04:36.165
Six capitalization out
84
00:04:36.185 --> 00:04:40.165
of more than 3000 companies listed in the nasdaq,
85
00:04:40.675 --> 00:04:44.565
it's 0.2% of the number of firms.
86
00:04:45.225 --> 00:04:46.605
The total capitalization
87
00:04:46.605 --> 00:04:51.405
of NASDAQ is more than 26,000 billions
88
00:04:51.505 --> 00:04:53.405
of dollars trillion dollars.
89
00:04:54.025 --> 00:04:57.125
Now, these six largest capitalized agents, they represent
90
00:04:57.585 --> 00:05:01.645
around 17 trillion, which is 63%.
91
00:05:02.465 --> 00:05:05.085
Now, if you, an extremely negligible number
92
00:05:05.085 --> 00:05:08.045
of firms represent about two thirds of the value,
93
00:05:08.505 --> 00:05:11.965
you understand that we are no more in a normal distribution
94
00:05:11.985 --> 00:05:15.485
in a Ian world, extremes have
95
00:05:16.005 --> 00:05:17.765
replaced the normal distribution.
96
00:05:18.585 --> 00:05:20.725
Now, if you look at the theory of corporate finance,
97
00:05:20.725 --> 00:05:22.565
which is named the portfolio theory,
98
00:05:22.875 --> 00:05:24.565
when you look at the models thanks
99
00:05:24.565 --> 00:05:27.605
to which you can make some option valuation, blackshaw
100
00:05:27.765 --> 00:05:31.565
and so on so forth, it's based about log normal distribution
101
00:05:31.565 --> 00:05:34.245
of returns in the portfolio theory.
102
00:05:34.625 --> 00:05:37.325
Now it's wrong. It does not work anymore.
103
00:05:37.915 --> 00:05:40.365
It's zero multiplied by infinity.
104
00:05:40.665 --> 00:05:44.485
So the portfolio theory needs a little bit of refurbishment
105
00:05:45.025 --> 00:05:47.805
or maybe complete reconstruction.
106
00:05:48.025 --> 00:05:51.365
Who has his sixth largest capitalization on nasdaq?
107
00:05:51.585 --> 00:05:56.445
The first one up to January 27 is Nvidia.
108
00:05:57.145 --> 00:06:00.965
Nvidia generates revenues of about $100 billion
109
00:06:02.325 --> 00:06:03.965
multiplied by two each and every year.
110
00:06:04.185 --> 00:06:07.845
By the way, the EBITDA is more than 60 billion,
111
00:06:08.385 --> 00:06:10.805
so you understand that the return on sales is absolutely
112
00:06:10.805 --> 00:06:14.085
outstanding, but free cash flow is a big diners.
113
00:06:14.085 --> 00:06:17.845
Capital expenditures CapEx a billion. Why?
114
00:06:17.845 --> 00:06:20.565
Because they don't manufacture the market.
115
00:06:20.665 --> 00:06:22.725
Capitalization is stratospheric.
116
00:06:23.225 --> 00:06:26.645
The price to book market value of equity divided
117
00:06:26.825 --> 00:06:28.285
by the book equity, which you are in,
118
00:06:28.285 --> 00:06:30.285
and the balance sheet is more than 60
119
00:06:31.145 --> 00:06:35.485
and the enterprise value represents about 60 euro
120
00:06:35.785 --> 00:06:38.805
of EBDA, which is absolutely outstanding.
121
00:06:39.965 --> 00:06:42.285
Interestingly, apple is number two.
122
00:06:43.025 --> 00:06:46.325
The EBITDA is larger, 130 something.
123
00:06:46.865 --> 00:06:51.085
The revenues of the last 12 months is about 400 billion,
124
00:06:52.025 --> 00:06:54.525
and capital expenditures are not that big
125
00:06:54.525 --> 00:06:57.085
because everything is outsourced at Apple,
126
00:06:57.785 --> 00:06:59.565
the market cap is not that far
127
00:06:59.745 --> 00:07:01.845
and the price to book is about the same.
128
00:07:02.705 --> 00:07:06.005
Now, the enterprise value in EBITDA is 25,
129
00:07:06.175 --> 00:07:10.365
which is much less than 57 or 60 for Nvidia,
130
00:07:10.705 --> 00:07:13.925
and what is interesting is to observe the next four
131
00:07:14.685 --> 00:07:18.045
Microsoft, Google, Amazon and meta.
132
00:07:18.385 --> 00:07:22.325
The revenues are from 150, 160 for meta to
133
00:07:22.865 --> 00:07:26.605
620 billion for Amazon, which is getting closer
134
00:07:26.665 --> 00:07:27.725
and closer to Walmart.
135
00:07:28.025 --> 00:07:30.445
By the way. What is interesting is
136
00:07:30.445 --> 00:07:32.725
that the EBIT D rate is quite high,
137
00:07:33.145 --> 00:07:37.485
but for these companies from rank three to rank six,
138
00:07:37.705 --> 00:07:40.525
the capital expenditures are significantly higher
139
00:07:40.955 --> 00:07:44.645
because they invest in not only warehousing for Amazon
140
00:07:44.645 --> 00:07:49.045
but also in data centers, networks and so on so forth.
141
00:07:49.425 --> 00:07:51.925
As a consequence, what is interesting to observe is
142
00:07:51.925 --> 00:07:54.245
that the price to book is a bit less
143
00:07:54.475 --> 00:07:57.045
because this company is in much more equity
144
00:07:57.145 --> 00:07:58.205
to finance their growth,
145
00:07:58.225 --> 00:08:00.045
to finance their capital expenditures,
146
00:08:00.345 --> 00:08:02.325
but the price to book is less
147
00:08:02.425 --> 00:08:07.405
and the enterprise value over EBITDA is quote normal,
148
00:08:08.235 --> 00:08:10.965
it's about 20 or a bit less than 20.
149
00:08:11.785 --> 00:08:15.085
We all know companies which are not in the artificial
150
00:08:15.085 --> 00:08:18.565
intelligence world and whose enterprise value represents
151
00:08:18.565 --> 00:08:20.085
about 20 years of abit.
152
00:08:20.425 --> 00:08:22.525
Why? Because these companies are profitable,
153
00:08:22.835 --> 00:08:25.445
they are prospects for growth and so on, so forth.
154
00:08:26.225 --> 00:08:27.765
Of course, these companies are much
155
00:08:28.435 --> 00:08:31.765
smaller than the big companies, which I describe here,
156
00:08:31.985 --> 00:08:34.885
but the multiple is not completely stupid.
157
00:08:35.265 --> 00:08:38.325
The last point I would like to discuss with you is the beta,
158
00:08:38.585 --> 00:08:40.525
the systematic risk coefficient.
159
00:08:40.915 --> 00:08:42.885
It's quite high for Nvidia
160
00:08:42.945 --> 00:08:47.405
and what happened recently will probably make the beta go up
161
00:08:47.415 --> 00:08:50.205
again, but if you look at the other five,
162
00:08:50.545 --> 00:08:52.245
the beta is about to one.
163
00:08:53.185 --> 00:08:56.325
You know what? These companies are completely embedded in
164
00:08:56.605 --> 00:09:00.925
economic life and this is why the beta is quite correlated
165
00:09:01.075 --> 00:09:04.405
with macroeconomic conditions about one.
166
00:09:05.525 --> 00:09:07.325
Interestingly, if you make the parallel
167
00:09:07.325 --> 00:09:10.005
with the telecommunication business telecommunication
168
00:09:10.005 --> 00:09:12.765
equipment, the beta is much higher than
169
00:09:12.885 --> 00:09:14.405
telecommunication operators.
170
00:09:14.915 --> 00:09:16.125
It's quite the same story,
171
00:09:16.665 --> 00:09:18.245
but I would like that you remember
172
00:09:18.245 --> 00:09:21.725
that the enterprise value over EBITDA of 20 is
173
00:09:22.305 --> 00:09:23.605
not abnormal.
174
00:09:24.245 --> 00:09:26.805
A couple of conclusions about observing these figures.
175
00:09:27.235 --> 00:09:30.805
Huge operating margins, ebitda, return on sales,
176
00:09:31.565 --> 00:09:35.205
absolutely fantastic, very significant free cash flows.
177
00:09:35.205 --> 00:09:37.925
Sometimes the free cash flow is very close to the EBITDA
178
00:09:37.925 --> 00:09:39.085
because there's no CapEx.
179
00:09:39.275 --> 00:09:41.445
Sometimes it's much less than the ebitda,
180
00:09:41.545 --> 00:09:43.045
but it's still quite big.
181
00:09:43.585 --> 00:09:45.525
So you have very large companies
182
00:09:45.795 --> 00:09:50.285
with fantastic operating margins and strong free cash flows.
183
00:09:50.755 --> 00:09:54.845
It's a fantastic and concentrated financial power.
184
00:09:55.245 --> 00:09:57.685
A few additional comments about returning
185
00:09:57.685 --> 00:09:58.725
the cash to shareholders.
186
00:09:59.385 --> 00:10:03.445
The return on equity for Apple is 150, 160
187
00:10:04.125 --> 00:10:05.965
NVD is more than 100%.
188
00:10:06.825 --> 00:10:08.685
The return on equity is the rate
189
00:10:08.685 --> 00:10:09.925
of return for the shareholders.
190
00:10:10.185 --> 00:10:13.365
Now, if the return on equity is very high, it's quite normal
191
00:10:13.435 --> 00:10:15.405
that the price to book is very high
192
00:10:16.065 --> 00:10:17.925
and for these two companies, a price
193
00:10:17.925 --> 00:10:19.325
to book is exceeding 60,
194
00:10:20.105 --> 00:10:22.645
but the return on equity is not only the performance
195
00:10:22.645 --> 00:10:24.605
of the company, it is the ability to grow.
196
00:10:25.505 --> 00:10:28.525
You remember this concept of sustainable growth rates,
197
00:10:28.525 --> 00:10:31.765
the ability of the company to self-finance its growth
198
00:10:32.105 --> 00:10:36.365
and then the ability to grow is more than 100%
199
00:10:37.185 --> 00:10:38.565
if you don't take into account
200
00:10:38.565 --> 00:10:40.085
returning the cash to shareholders.
201
00:10:40.865 --> 00:10:43.045
Now, let's take these two companies one after the other.
202
00:10:43.495 --> 00:10:47.605
Apple revenues are owe by about 10% per year
203
00:10:48.155 --> 00:10:50.965
earning soft tax 2024
204
00:10:51.505 --> 00:10:53.685
$94 billion.
205
00:10:54.555 --> 00:10:57.925
Just to give you an idea about transforming that into cash,
206
00:10:58.025 --> 00:11:00.045
the company generates a free cash flow
207
00:11:00.045 --> 00:11:03.405
of $1 million every five minutes.
208
00:11:04.425 --> 00:11:06.365
So when you are profitable,
209
00:11:06.915 --> 00:11:09.765
when you are immensely profitable in generating a lot
210
00:11:09.765 --> 00:11:12.845
of cash and the gross rate is on lead 10%,
211
00:11:13.545 --> 00:11:14.845
you generate much more cash
212
00:11:14.845 --> 00:11:16.405
and what you need to final your growth,
213
00:11:16.435 --> 00:11:18.325
then you can return the cash to shareholders
214
00:11:18.945 --> 00:11:23.565
buy back 95 billion dividends, 15 billion,
215
00:11:23.745 --> 00:11:26.925
so we can return $110 billion.
216
00:11:27.625 --> 00:11:29.645
But what is interesting about Nvidia,
217
00:11:29.945 --> 00:11:33.125
Nvidia is doubling the revenues each and every year.
218
00:11:33.545 --> 00:11:36.525
So the revenue growth is 100% earnings
219
00:11:36.525 --> 00:11:41.325
after tax 2023 last, last 12 months and about 30 billion.
220
00:11:42.525 --> 00:11:46.165
But as a company does not invest CapEx 1 billion,
221
00:11:46.665 --> 00:11:49.045
the company can return the cash to the shareholders
222
00:11:49.965 --> 00:11:53.045
dividends and buybacks altogether during the last two years,
223
00:11:53.200 --> 00:11:55.925
they represent $20 billion.
224
00:11:56.505 --> 00:11:59.125
So you understand that these companies are extremely strong
225
00:11:59.385 --> 00:12:01.725
and they can give the money back to the shareholders.
226
00:12:01.915 --> 00:12:04.165
What are shareholders going to do with this money?
227
00:12:05.205 --> 00:12:07.085
Reinvest in artificial intelligence.
228
00:12:07.265 --> 00:12:10.125
So you have plenty your financial resources ready
229
00:12:10.145 --> 00:12:12.645
to invest in the development of this industry
230
00:12:13.065 --> 00:12:15.765
and generated by the leaders of this industry.
231
00:12:16.065 --> 00:12:19.405
That's a huge value, but sometimes the value is fraud.
232
00:12:20.225 --> 00:12:22.925
We can take the example of what happened to Google
233
00:12:23.665 --> 00:12:26.285
barred the artificial intelligence application
234
00:12:26.285 --> 00:12:29.085
of Google in February, 2023.
235
00:12:29.785 --> 00:12:32.325
The company in the middle of a presentation
236
00:12:32.945 --> 00:12:35.405
is providing an inaccurate information.
237
00:12:36.345 --> 00:12:39.525
Rer is in a room and immediately circulates a mistake.
238
00:12:39.675 --> 00:12:43.165
What immediately happens in the stock market time magazine
239
00:12:43.175 --> 00:12:46.125
title an error just cost Google
240
00:12:46.425 --> 00:12:48.445
$100 billion.
241
00:12:49.105 --> 00:12:51.725
So this figure is absolutely stratospheric,
242
00:12:52.305 --> 00:12:54.085
but it's extremely temporary.
243
00:12:54.465 --> 00:12:57.605
If you look at the graph, the evolution of alphabet
244
00:12:57.605 --> 00:13:00.805
of Google throughout the years, you understand that it's up,
245
00:13:00.875 --> 00:13:04.765
it's down and there's a blip somewhere on the screen just
246
00:13:04.865 --> 00:13:07.765
for a few days, a few weeks, and then it's back.
247
00:13:08.115 --> 00:13:10.205
It's back up. The trend is there.
248
00:13:11.345 --> 00:13:14.525
If you remember Facebook now meta
249
00:13:14.875 --> 00:13:16.685
with Cambridge Analytica.
250
00:13:16.995 --> 00:13:20.845
Cambridge Analytica was very much an issue for Facebook.
251
00:13:21.345 --> 00:13:22.485
The third price went down.
252
00:13:23.055 --> 00:13:26.685
Three weeks later it was up and it started going up again.
253
00:13:27.185 --> 00:13:29.005
So it's just an accident.
254
00:13:29.355 --> 00:13:33.165
It's a $100 billion temporary accident.
255
00:13:33.705 --> 00:13:37.085
Now, AI is not limited to the big six, which I mentioned.
256
00:13:37.875 --> 00:13:40.525
Open AI of course is in the middle of the picture
257
00:13:41.265 --> 00:13:44.685
and tropic x ai, mytral ai,
258
00:13:45.365 --> 00:13:48.085
a few mobility cases and Reddit case.
259
00:13:48.395 --> 00:13:50.485
It's what I'm going to discuss now with you.
260
00:13:51.495 --> 00:13:52.925
Let's start with open ai.
261
00:13:53.305 --> 00:13:57.765
OpenAI, very well known Chad g PT founded in 2015
262
00:13:57.945 --> 00:14:01.925
by Sam Altman, who a very good friend of Sam Alman,
263
00:14:01.925 --> 00:14:04.725
who's name is Elon Musk and a few others.
264
00:14:05.625 --> 00:14:07.285
It seems that Sam Atman
265
00:14:07.285 --> 00:14:09.565
and Elon Musk are no more friends anymore,
266
00:14:09.975 --> 00:14:13.765
which happen in uh, in number of cases in this industry.
267
00:14:14.465 --> 00:14:18.405
Why did they create open AI for the benefit of humankind?
268
00:14:18.945 --> 00:14:21.125
The free cash flow is never that far from that.
269
00:14:21.705 --> 00:14:24.965
Now, at that time, Sam Atman is the president
270
00:14:25.065 --> 00:14:27.885
of an outstanding accelerator based in California.
271
00:14:28.025 --> 00:14:29.685
His name is Y Combinator.
272
00:14:30.345 --> 00:14:35.085
In 2019, he leaves Y Combinator to be 100% CEO
273
00:14:35.305 --> 00:14:36.525
of open ai.
274
00:14:37.355 --> 00:14:39.845
He's going to be host stood by the board
275
00:14:40.345 --> 00:14:42.725
and then back in the board, right?
276
00:14:43.025 --> 00:14:46.365
And then a few people are going to be out in the board
277
00:14:46.545 --> 00:14:50.205
and top managers, top people at OpenAI, three co-founders,
278
00:14:51.165 --> 00:14:53.925
C-S-O-C-T-O security r and d manager
279
00:14:54.425 --> 00:14:57.925
and a number of people are living open ai,
280
00:14:57.935 --> 00:15:01.085
which is definitely an issue for the company in the future
281
00:15:01.785 --> 00:15:04.245
as far as some money is concerned is still there
282
00:15:04.825 --> 00:15:06.965
and what's happened to the value of the company?
283
00:15:07.275 --> 00:15:11.125
Well, in 2023, Microsoft, the long term financial
284
00:15:11.705 --> 00:15:15.565
and business partner and technological partner of open AI
285
00:15:16.045 --> 00:15:17.645
provided $10 billion.
286
00:15:17.955 --> 00:15:20.645
This amount has to be compared with the primo valuation.
287
00:15:20.645 --> 00:15:24.325
So the value before you receive the cash of 19 billion.
288
00:15:24.985 --> 00:15:28.605
So post money, the value is 19%, which is 29,
289
00:15:29.055 --> 00:15:33.045
about 30 billion a year later, exactly 18 months later.
290
00:15:33.185 --> 00:15:36.685
In 2024, an additional cash injection
291
00:15:36.945 --> 00:15:40.245
of $6.6 billion out of a premium
292
00:15:40.625 --> 00:15:42.165
of 150.
293
00:15:42.825 --> 00:15:45.405
Now the 150 has to be compared with 30,
294
00:15:45.905 --> 00:15:48.285
and so the value of the company was multiplied
295
00:15:48.285 --> 00:15:50.645
by five in just 18 months.
296
00:15:51.505 --> 00:15:54.485
So we are not exactly in the same valuation process
297
00:15:54.785 --> 00:15:57.885
as normal companies and normal industries.
298
00:15:58.465 --> 00:16:00.605
Tropic is a company which was founded
299
00:16:00.745 --> 00:16:02.845
by former open AI people
300
00:16:02.845 --> 00:16:05.325
because a number of them left the company.
301
00:16:05.865 --> 00:16:08.365
The company was founded in 2021.
302
00:16:08.985 --> 00:16:12.525
The last valuation of the company series D,
303
00:16:13.165 --> 00:16:16.405
February, 2024 mentions about 17
304
00:16:16.505 --> 00:16:18.325
to $18 billion.
305
00:16:18.925 --> 00:16:21.525
A brother and a sister created the company.
306
00:16:22.145 --> 00:16:25.365
It seems that they signed a big contract with Amazon
307
00:16:25.425 --> 00:16:27.445
to provide service and so on so forth,
308
00:16:27.465 --> 00:16:31.925
and they received some financing interestingly from Google,
309
00:16:32.515 --> 00:16:35.805
from the speculative fund, which was created
310
00:16:36.025 --> 00:16:37.845
by SoftBank Vision Fund.
311
00:16:38.235 --> 00:16:40.525
Keep in mind, SoftBank, we're going
312
00:16:40.525 --> 00:16:42.325
to talk about the capital later
313
00:16:43.065 --> 00:16:45.765
and the financing was about $4 billion.
314
00:16:46.585 --> 00:16:48.725
The valuation are stratospheric,
315
00:16:48.745 --> 00:16:50.805
but it's just paper valuation.
316
00:16:51.145 --> 00:16:53.245
The MONA is available in cash.
317
00:16:53.675 --> 00:16:57.085
Elon Musk is no more friend of uh, Sam man.
318
00:16:57.425 --> 00:17:01.205
He left, uh, open ai. He said, we have to stop everything.
319
00:17:01.225 --> 00:17:02.405
Please stop everything
320
00:17:02.405 --> 00:17:05.005
because in the meantime, I'm creating my own company.
321
00:17:05.465 --> 00:17:10.165
His name is XAIX AI because of SpaceX,
322
00:17:10.165 --> 00:17:12.965
because of Twitter becoming X and so on.
323
00:17:13.105 --> 00:17:16.605
The company is founded in July, 2023.
324
00:17:17.425 --> 00:17:20.885
The name of the application is extremely poetic gr
325
00:17:21.785 --> 00:17:23.565
and in May, 2024,
326
00:17:24.025 --> 00:17:27.645
the company leads Monet $6 billion out
327
00:17:27.645 --> 00:17:32.565
of a pre month valuation of 18 billion who participated
328
00:17:32.665 --> 00:17:36.005
to the equity issue, the usual suspect, Sequoia
329
00:17:37.955 --> 00:17:40.645
plus a few sovereign funds and so on so forth.
330
00:17:41.025 --> 00:17:44.685
It was in May Primo 18 in December,
331
00:17:45.065 --> 00:17:49.325
we need an additional $6 billion, PRI Monet, 44 billion.
332
00:17:49.985 --> 00:17:53.965
So again, the value is up, but the value is not cash.
333
00:17:54.315 --> 00:17:56.205
What is injected in these companies
334
00:17:56.205 --> 00:17:58.205
to pay the salaries is cash.
335
00:17:58.425 --> 00:18:00.965
In Europe. In France, there's a company which seems
336
00:18:00.965 --> 00:18:03.565
to be quite promising in the same industry name is
337
00:18:03.845 --> 00:18:05.005
Mistrial ai.
338
00:18:05.745 --> 00:18:08.005
We have a contract with Microsoft
339
00:18:08.345 --> 00:18:12.165
and they have a nice credibility in terms of technology.
340
00:18:12.225 --> 00:18:16.485
The company was created in 20 23, 3 founders coming from
341
00:18:16.485 --> 00:18:20.125
Google DeepMind and from Meta Facebook,
342
00:18:20.665 --> 00:18:24.005
but in order to raise just $1.1 billion,
343
00:18:24.585 --> 00:18:27.805
the company needed 40 investors, which is
344
00:18:28.485 --> 00:18:31.365
complete different context in terms of venture capital
345
00:18:31.945 --> 00:18:34.485
series B in June, 2024.
346
00:18:35.025 --> 00:18:37.805
Now, premium evaluation is 5.2 billion.
347
00:18:37.905 --> 00:18:40.285
The capital is raising half a billion.
348
00:18:41.025 --> 00:18:44.885
The name of the application is Lucia Lecha,
349
00:18:45.585 --> 00:18:47.085
so it's about chat,
350
00:18:47.265 --> 00:18:51.805
but Lucia is a cartoon humoristic cartoon in Europe,
351
00:18:51.855 --> 00:18:53.205
which is extremely well known
352
00:18:53.265 --> 00:18:55.325
and very much appreciated by the way.
353
00:18:55.665 --> 00:18:56.885
Now if you look at
354
00:18:56.955 --> 00:19:01.725
what Crunchbase very serious website is giving us a list of
355
00:19:02.265 --> 00:19:06.085
who received more than $1 billion in the last two years.
356
00:19:06.105 --> 00:19:07.605
23 companies,
357
00:19:08.185 --> 00:19:10.765
of course the usual suspects I already mentioned,
358
00:19:11.065 --> 00:19:13.325
but two companies in the mobility business
359
00:19:13.385 --> 00:19:15.405
and it's interesting to spend a minute on that.
360
00:19:15.505 --> 00:19:20.365
The first one is Waymo Ymo is Google mobility activity
361
00:19:20.665 --> 00:19:24.405
and the company received in October 20, 24, 5
362
00:19:24.455 --> 00:19:27.725
$0.6 billion to develop autonomous driving
363
00:19:27.985 --> 00:19:29.045
and so on so forth.
364
00:19:29.825 --> 00:19:32.565
Cruise is also a very interesting case.
365
00:19:32.955 --> 00:19:36.805
It's about self-driving vehicles, it's about robo taxis.
366
00:19:36.805 --> 00:19:40.005
The company received plenty of money from GM
367
00:19:40.665 --> 00:19:42.965
and SoftBank Vision Fund,
368
00:19:43.545 --> 00:19:47.085
but now GM is a controlling interest in growth.
369
00:19:47.285 --> 00:19:48.525
I will describe a bit later.
370
00:19:49.145 --> 00:19:52.085
The company received 2.2 in 2022
371
00:19:52.585 --> 00:19:55.725
and 850 million in 2024
372
00:19:56.605 --> 00:19:58.125
provided by General Motor.
373
00:19:58.265 --> 00:20:01.165
It is a total of $16 billion
374
00:20:01.585 --> 00:20:03.565
of funds received from investors,
375
00:20:03.955 --> 00:20:06.565
including the largest one general models.
376
00:20:07.065 --> 00:20:09.165
Now, again, the value is fragile.
377
00:20:10.345 --> 00:20:12.525
GM grew, it's not a great investment.
378
00:20:12.915 --> 00:20:16.285
They developed a business fleet of Robax.
379
00:20:17.145 --> 00:20:20.405
The company took off in 2018 with GM
380
00:20:20.785 --> 00:20:22.485
and SoftBank Vision Fund.
381
00:20:22.565 --> 00:20:24.765
SoftBank provided $2 billion
382
00:20:25.705 --> 00:20:29.605
and GM bought the Southbank Vision fund stocks in
383
00:20:29.605 --> 00:20:30.805
2022,
384
00:20:31.345 --> 00:20:34.005
and now the company has almost the whole control,
385
00:20:34.105 --> 00:20:38.965
the entire control of uh, CRO revenues generated
386
00:20:39.025 --> 00:20:40.645
by the company throughout the years.
387
00:20:40.715 --> 00:20:44.605
It's about half a billion, but GM invested $10 billion
388
00:20:44.945 --> 00:20:46.765
and it's not successful at all.
389
00:20:47.315 --> 00:20:49.725
This is why GM recently said, we're going
390
00:20:49.725 --> 00:20:51.325
to stop everything about that.
391
00:20:51.825 --> 00:20:54.325
We are going to stop the robo taxi business.
392
00:20:54.505 --> 00:20:55.965
We are going to take the technology
393
00:20:56.025 --> 00:20:59.445
and bring the technology back to GM for personal vehicles,
394
00:20:59.755 --> 00:21:01.725
driving assistance and so on so forth.
395
00:21:01.945 --> 00:21:05.765
And instead of costing 10 billion wasted money, we are going
396
00:21:05.765 --> 00:21:07.685
to invest about a billion per year.
397
00:21:08.425 --> 00:21:12.725
So GM had bought the SoftBank Vision Fund shares,
398
00:21:13.155 --> 00:21:14.445
they had 90%.
399
00:21:14.545 --> 00:21:17.405
Now they have an option to go to 97%, they're going
400
00:21:17.405 --> 00:21:19.605
to buy back all the share end of the story,
401
00:21:20.145 --> 00:21:22.205
and Cruise is not a success.
402
00:21:22.265 --> 00:21:25.405
So it's not only about successes in this industry
403
00:21:25.785 --> 00:21:27.565
and sometimes life stops.
404
00:21:27.905 --> 00:21:30.765
Reddit is the last case I would like to discuss with you.
405
00:21:31.225 --> 00:21:34.685
You can't find on the academy an educational film dated
406
00:21:35.215 --> 00:21:38.245
April, 2024, not that far ago,
407
00:21:38.745 --> 00:21:41.725
and it was about E-S-G-E-S-G,
408
00:21:41.725 --> 00:21:46.365
which potentially creates value thanks to the evolution
409
00:21:46.425 --> 00:21:47.525
of ESG.
410
00:21:47.585 --> 00:21:50.645
At Reddit, the company could be listed
411
00:21:50.865 --> 00:21:52.845
and so it created value for its shareholders.
412
00:21:53.425 --> 00:21:55.645
Reddit is a community of communities,
413
00:21:55.925 --> 00:22:00.485
a sub Reddit company created in 2005.
414
00:22:01.145 --> 00:22:05.485
In 2014, series B, the company is raising $50 million.
415
00:22:06.115 --> 00:22:09.245
Most of the money comes from Sam Atman
416
00:22:09.585 --> 00:22:11.285
and was a good investment for Mr.
417
00:22:11.495 --> 00:22:12.805
Atman because the value
418
00:22:13.025 --> 00:22:16.565
of his equity stake in Reddi is supposedly about
419
00:22:16.705 --> 00:22:17.885
$600 million.
420
00:22:18.095 --> 00:22:21.285
Today, thanks to becoming ESG friendly,
421
00:22:21.625 --> 00:22:25.405
Reddit became public company could go for an IPO.
422
00:22:25.505 --> 00:22:29.805
The IPO started at $8 billion. Now it's about $11 billion.
423
00:22:29.945 --> 00:22:33.565
But the question is what is the main asset of Reddi?
424
00:22:33.655 --> 00:22:37.085
Reddi is publishing is communicating on its strategy
425
00:22:37.785 --> 00:22:39.725
and the evolution of the business model.
426
00:22:39.935 --> 00:22:44.045
First, it's about advertising, segmentation, positioning,
427
00:22:44.275 --> 00:22:47.045
data return investment for those people
428
00:22:47.265 --> 00:22:48.485
who need advertising.
429
00:22:48.775 --> 00:22:52.165
Great second layer data licensing,
430
00:22:52.945 --> 00:22:56.565
all these people discussing, interacting in the communities
431
00:22:56.625 --> 00:22:58.085
and in the subreddits.
432
00:22:58.715 --> 00:23:03.125
It's a fantastic playground to train your AI applications.
433
00:23:03.705 --> 00:23:07.565
And so at the end of the day, Reddit is making money selling
434
00:23:07.785 --> 00:23:10.725
to big companies the usual suspects.
435
00:23:10.765 --> 00:23:15.045
I already imagine selling the access to the subreddits so
436
00:23:15.045 --> 00:23:16.765
that they can train the application
437
00:23:17.145 --> 00:23:18.525
and you know how much it costs.
438
00:23:18.705 --> 00:23:19.845
You know how strategy
439
00:23:19.985 --> 00:23:23.045
and critical this activity is for those people,
440
00:23:23.215 --> 00:23:26.045
those companies providing the AI applications.
441
00:23:26.385 --> 00:23:29.165
Now, I provided some information about large
442
00:23:29.505 --> 00:23:32.565
and less, large, but key players in this industry.
443
00:23:33.055 --> 00:23:36.165
There are plenty of remaining issues.
444
00:23:36.765 --> 00:23:39.325
I said I'm not going to discuss threats on ethics,
445
00:23:39.705 --> 00:23:41.205
but it's nice to mention
446
00:23:41.435 --> 00:23:44.485
that these companies are consuming a lot of energy.
447
00:23:44.995 --> 00:23:46.845
They're issuing plenty of carbon.
448
00:23:47.515 --> 00:23:51.605
It's a oligo poly, an extremely restricted oligo poly.
449
00:23:52.305 --> 00:23:55.565
Not many companies with fantastic resources.
450
00:23:55.675 --> 00:23:59.405
Valuation and cash sovereignty is a little bit
451
00:23:59.405 --> 00:24:01.925
of problem consumption of metals
452
00:24:02.385 --> 00:24:05.285
and related with that, the recycling issue,
453
00:24:06.145 --> 00:24:09.365
the data centers, they consume huge space,
454
00:24:09.615 --> 00:24:11.685
which is not available to do anything else.
455
00:24:12.295 --> 00:24:14.605
These companies, they have plenty of cash
456
00:24:14.705 --> 00:24:16.525
and they attract competencies
457
00:24:17.345 --> 00:24:20.325
and technical competencies are quite scarce,
458
00:24:20.665 --> 00:24:22.685
but as they have plenty of money,
459
00:24:22.795 --> 00:24:24.605
they can offer extremely attractive
460
00:24:25.005 --> 00:24:26.165
remuneration for these people.
461
00:24:26.635 --> 00:24:29.605
This competence is now working for artificial intelligence,
462
00:24:29.665 --> 00:24:31.925
are not available for the classical
463
00:24:32.185 --> 00:24:33.845
and traditional industries.
464
00:24:34.275 --> 00:24:37.525
This is the advantage of being strong, powerful,
465
00:24:37.745 --> 00:24:39.525
and full of financial resources.
466
00:24:39.895 --> 00:24:42.805
There are also plenty of issues about cyber security,
467
00:24:43.545 --> 00:24:44.765
et cetera, et cetera.
468
00:24:45.085 --> 00:24:48.685
A couple of very recent examples to conclude this film.
469
00:24:49.305 --> 00:24:50.525
You remember in the nineties,
470
00:24:50.805 --> 00:24:54.005
Stargate Stargate was a science fiction film.
471
00:24:54.745 --> 00:24:59.325
Now 20, 25, January the 22nd of January,
472
00:24:59.835 --> 00:25:02.965
what is announced by open AI here on X,
473
00:25:03.585 --> 00:25:06.165
we have created the Stargate project.
474
00:25:06.275 --> 00:25:07.685
It's going to be joint venture
475
00:25:07.795 --> 00:25:11.365
with a few actors including open ai, of course,
476
00:25:11.715 --> 00:25:13.765
including SoftBank, Mai
477
00:25:13.765 --> 00:25:16.725
or Csan, including Larry Ellison Rockle,
478
00:25:17.025 --> 00:25:18.085
and we are going
479
00:25:18.085 --> 00:25:22.925
to invest $500 billion in artificial intelligence
480
00:25:23.435 --> 00:25:24.965
over the next four years.
481
00:25:25.545 --> 00:25:27.325
That's absolutely outstanding.
482
00:25:28.285 --> 00:25:32.485
I already did discussed Masayoshi son investment cycle
483
00:25:33.025 --> 00:25:36.685
at SoftBank, SoftBank Vision Fund and so on.
484
00:25:37.385 --> 00:25:40.165
Now, the person is very much involved in ai.
485
00:25:40.945 --> 00:25:44.365
The opportunity I had to discuss about some investments
486
00:25:44.365 --> 00:25:47.965
where about we work, which was not exactly successful,
487
00:25:48.505 --> 00:25:51.205
but Massa Oran says we have to bring
488
00:25:51.775 --> 00:25:53.645
brain computer to life.
489
00:25:54.105 --> 00:25:55.445
So we have to be able
490
00:25:55.465 --> 00:25:59.965
to transform the biological brain into a set of algorithm
491
00:26:00.725 --> 00:26:02.365
combined with microchips.
492
00:26:02.575 --> 00:26:05.085
Those this is brain, computer to life.
493
00:26:05.555 --> 00:26:07.245
This is a fantastic dream.
494
00:26:07.735 --> 00:26:10.285
Successful or not, we are going to see that in the future.
495
00:26:10.785 --> 00:26:14.605
But what is interesting that SoftBank was in a number
496
00:26:14.665 --> 00:26:16.045
of different businesses
497
00:26:16.505 --> 00:26:19.725
and now they are a little bit catching up about ai.
498
00:26:20.435 --> 00:26:25.125
They had a very good start buying ARM holdings, the company,
499
00:26:25.335 --> 00:26:27.485
which Nvidia wanted to buy by the way,
500
00:26:27.625 --> 00:26:29.365
and the antitrust author said,
501
00:26:29.735 --> 00:26:32.685
don't even think about it when they bought arm,
502
00:26:32.905 --> 00:26:37.485
it is in 2060 and they pay $31 billion for the acquisition.
503
00:26:37.485 --> 00:26:40.885
2023, the company goes public at the value,
504
00:26:40.885 --> 00:26:42.725
which is 54 billion.
505
00:26:42.955 --> 00:26:45.885
Billion. That's a good deal, and today it's 155.
506
00:26:46.025 --> 00:26:47.125
That's a good investment.
507
00:26:48.105 --> 00:26:51.685
Now, the very best investment ever made by Mr.
508
00:26:51.935 --> 00:26:55.445
Masai Oran in the tech business is Alibaba.
509
00:26:56.105 --> 00:26:57.885
He invested $100 million,
510
00:26:57.885 --> 00:27:00.565
which was transformed eventually into tens
511
00:27:00.585 --> 00:27:01.925
of billions of dollars.
512
00:27:02.305 --> 00:27:04.205
The second one is Nvidia.
513
00:27:04.625 --> 00:27:08.045
Congratulations because he invested $4 billion in
514
00:27:08.105 --> 00:27:11.965
May, 2017 when the market cap was about
515
00:27:12.225 --> 00:27:13.445
$80 billion.
516
00:27:14.025 --> 00:27:15.885
It was already a quite high figure,
517
00:27:15.985 --> 00:27:18.685
but it has been modified by 40 since then.
518
00:27:19.305 --> 00:27:22.445
He invested in cruise. You remember the GM story?
519
00:27:22.905 --> 00:27:26.485
He invested two point something billion in 2018
520
00:27:26.585 --> 00:27:29.965
and sold in 2022 for two point billion something
521
00:27:30.715 --> 00:27:34.005
plus minus zero per cm open AI yet,
522
00:27:34.065 --> 00:27:37.845
but quite late in October, 2024.
523
00:27:38.625 --> 00:27:42.805
So NVDI is definitely a very high success arm,
524
00:27:43.065 --> 00:27:44.845
is quite good profitability.
525
00:27:45.305 --> 00:27:49.645
And so SoftBank Vision Fund is a very significant
526
00:27:49.895 --> 00:27:51.125
actor in ai.
527
00:27:52.035 --> 00:27:54.325
This is why, and you see that on the picture.
528
00:27:54.785 --> 00:27:57.925
The president of the United States is welcoming
529
00:27:59.115 --> 00:28:02.445
maan together with Larry Ellison and Samba.
530
00:28:02.985 --> 00:28:06.085
The joint venture is going to be funded by OpenAI,
531
00:28:06.445 --> 00:28:10.285
19 billion, soft bank, 19 billion or rl.
532
00:28:10.285 --> 00:28:13.005
We don't know exactly how much plus one or two additional,
533
00:28:13.705 --> 00:28:15.045
and at the end of the day it's going
534
00:28:15.365 --> 00:28:19.165
to be $500 billion over the next four years.
535
00:28:19.865 --> 00:28:22.125
But on this picture, there are two people
536
00:28:22.345 --> 00:28:24.165
who are absolutely missing.
537
00:28:24.945 --> 00:28:27.165
The first one is Microsoft.
538
00:28:28.165 --> 00:28:30.845
Microsoft originally was a financial partner.
539
00:28:30.995 --> 00:28:34.565
They provide a lot of funding for open ai.
540
00:28:35.065 --> 00:28:37.005
Now they say, oh, this is going
541
00:28:37.005 --> 00:28:40.445
to be a very strong technology partner together with Arm,
542
00:28:41.205 --> 00:28:45.245
SoftBank, Microsoft, Nvidia, or Rackl, et cetera, et cetera.
543
00:28:46.085 --> 00:28:48.325
Microsoft is no more a financial partner.
544
00:28:48.615 --> 00:28:50.805
Maybe it's a technological called partner.
545
00:28:51.545 --> 00:28:54.485
So it's now tier two in the process.
546
00:28:55.825 --> 00:28:58.765
The other one who is missing is Elon Musk.
547
00:28:59.475 --> 00:29:03.805
Elon Musk co-created open AI with Sam Atman.
548
00:29:03.945 --> 00:29:06.085
Now they are not exactly friends
549
00:29:06.745 --> 00:29:09.805
and Elon Musk brought open AI
550
00:29:10.105 --> 00:29:13.365
and Sam Atman to the court now adding
551
00:29:13.965 --> 00:29:15.765
Microsoft antitrust claims.
552
00:29:15.945 --> 00:29:19.245
We are November, 2024 and definitely Mr.
553
00:29:19.345 --> 00:29:22.285
Maersk, who seems to be good friend with the president
554
00:29:22.285 --> 00:29:24.405
of the United States is not a good friend
555
00:29:24.825 --> 00:29:26.285
of Stargate Project.
556
00:29:26.705 --> 00:29:30.365
The second event, which extremely recently happened,
557
00:29:30.475 --> 00:29:35.165
January 27, is deep seek a Chinese startup
558
00:29:35.785 --> 00:29:37.725
and they say, okay, we are doing good job.
559
00:29:37.905 --> 00:29:41.325
We are proposing an application which is almost as good
560
00:29:41.425 --> 00:29:45.805
as Chad G, but we didn't pass same price as Chad G.
561
00:29:46.545 --> 00:29:51.085
In fact, developing the application cost about 3% of
562
00:29:51.085 --> 00:29:53.525
what it cost to open AI to develop Chad,
563
00:29:53.565 --> 00:29:55.405
GPT 1, 2, 3, 4, et cetera, et cetera.
564
00:29:55.465 --> 00:29:59.325
And the consequence, what happened to Nvidia stock price,
565
00:29:59.705 --> 00:30:01.605
it went down dramatically.
566
00:30:02.065 --> 00:30:05.205
The company lost $600 billion
567
00:30:05.225 --> 00:30:07.365
of market cap in one day,
568
00:30:08.345 --> 00:30:12.205
$1 trillion evaporated in one day as a consequence
569
00:30:12.225 --> 00:30:13.685
of Dipe announcement
570
00:30:13.785 --> 00:30:17.405
and what the market observe about the efficiency
571
00:30:17.425 --> 00:30:18.485
of the application.
572
00:30:18.865 --> 00:30:23.205
Why? Because very likely all these investments are probably
573
00:30:23.915 --> 00:30:27.005
overestimated in terms of value creation.
574
00:30:27.305 --> 00:30:29.205
And there are a few guys here over there
575
00:30:29.825 --> 00:30:33.845
who can do a quite good job spending much less than all
576
00:30:33.845 --> 00:30:36.685
these big companies, which I already mentioned.
577
00:30:37.195 --> 00:30:40.165
Consequence Nvidia Verage down together
578
00:30:40.195 --> 00:30:42.285
with some other companies.
579
00:30:43.465 --> 00:30:47.245
Now the Chinese AI business looks a little bit better,
580
00:30:47.345 --> 00:30:48.805
and this is why, for example,
581
00:30:48.965 --> 00:30:51.125
a company whose name is Merit Interactive.
582
00:30:51.425 --> 00:30:53.965
So the stock price is skyrocketing
583
00:30:54.515 --> 00:30:58.925
because the one man's misfortune is outta the man's gain in
584
00:30:58.925 --> 00:31:00.685
finance with him that the portfolio
585
00:31:01.145 --> 00:31:04.045
and diversifying the specific risk.
586
00:31:04.425 --> 00:31:05.965
So if these guys are able
587
00:31:05.965 --> 00:31:07.885
to do a good job spending much less,
588
00:31:07.985 --> 00:31:09.965
the value is much higher stock cross.
589
00:31:10.075 --> 00:31:12.765
He's plus 20% in one day,
590
00:31:13.455 --> 00:31:15.445
which is absolutely outstanding
591
00:31:15.905 --> 00:31:19.925
and drives me to the first of my four conclusions.
592
00:31:20.225 --> 00:31:23.165
The first one is when you watch this video,
593
00:31:23.715 --> 00:31:26.445
unfortunately it's going to be obsolete
594
00:31:26.475 --> 00:31:28.925
because life moves on very quickly
595
00:31:29.385 --> 00:31:31.685
and from the moment I recorded the video
596
00:31:31.985 --> 00:31:33.925
and the moment you watched it, plenty
597
00:31:33.985 --> 00:31:35.325
of things will have happened,
598
00:31:35.615 --> 00:31:37.845
which makes this video a little bit obsolete.
599
00:31:37.865 --> 00:31:41.365
At least it'll give you some information about the past
600
00:31:41.915 --> 00:31:44.525
hopefully second conclusion,
601
00:31:44.525 --> 00:31:46.125
which is a little bit more serious.
602
00:31:46.465 --> 00:31:51.245
The financial power is hugely concentrated in an extremely
603
00:31:51.245 --> 00:31:52.805
limited number of hands.
604
00:31:53.605 --> 00:31:57.045
I described the six big players plus a few additional.
605
00:31:57.105 --> 00:31:58.845
You understand their financial power.
606
00:31:59.195 --> 00:32:01.485
It's a oligopoly, it's a strong oligopoly
607
00:32:01.485 --> 00:32:05.765
with outstanding operating margins, huge free cash flows,
608
00:32:05.765 --> 00:32:09.645
fantastic power to develop the business and take the power
609
00:32:10.025 --> 00:32:11.045
and take the control.
610
00:32:11.745 --> 00:32:15.885
Hugely concentrated power, certainly not goals curve.
611
00:32:16.385 --> 00:32:18.525
And now we have to reconsider, we have
612
00:32:18.525 --> 00:32:21.165
to rebuild the corporate finance theory.
613
00:32:21.555 --> 00:32:24.925
It's no more, it's normal, a log normal distribution.
614
00:32:25.065 --> 00:32:27.965
It is zero multiplied by infinity.
615
00:32:28.275 --> 00:32:29.685
It's completely different context.
616
00:32:30.105 --> 00:32:31.885
We have to start from scratch again
617
00:32:32.705 --> 00:32:35.805
and we have observed friends Elon Musk
618
00:32:35.955 --> 00:32:38.885
with Sam Atman, Microsoft yesterday.
619
00:32:39.485 --> 00:32:41.645
Microsoft Today. And about friendship,
620
00:32:41.665 --> 00:32:44.725
it reminds me is a very well known Raider on Wall Street,
621
00:32:44.915 --> 00:32:48.285
Carl Aiken, who used to say about friendship.
622
00:32:48.385 --> 00:32:52.005
If you want a friend, you should buy yourself a dog.
623
00:32:52.195 --> 00:32:55.285
It's about friendship in artificial intelligence,
624
00:32:55.545 --> 00:32:57.525
in finance, and on Wall Street.
625
00:32:58.055 --> 00:32:58.885
Thank you very much.
Hello on.
Welcome to this educational field whose objective is to provide you with information and financial insight into an economic sector, which is definitely at the forefront of business news.
Artificial intelligence, the man you see in the picture is named Jensen Huang.
He is the founder of a company which is now very well known whose name is Nvidia.
Nvidia was created by him in 1993.
He is 30 years old.
He's an experienced engineer, is very knowledgeable about all these technologies around microchips today.
I mean, before January 27, Nvidia was the largest market capitalization in the world with about $3,600 billion of market value on a nasdaq.
Now after some equity issues, of course Mr.
Huang has been a little bit diluted and it seems that he holds 3.6% of his shares, but 3.6% of this amount money represents about $130 billion and positions.
Mr.
Huang at the number 10 of the Forbes list of billionaires, which is not bad news for him.
Now, when you observe the evolution of Apple and Microsoft Giants in this business, in this industry compared with Nvidia over the last four or five years, what are the conclusions? Apple is up by 240%, which is great.
Microsoft is up by 180%, which is great.
Nvidia is up by 2700%, which is outstanding and rarely observed the stock market.
Now obviously Nvidia is not alone the planet in this industry.
You have the gamma former G, but you also have Chad D, pity, open ai, xai and Tropic menstrual AI and so on and so forth.
My objective in this field is not to describe the entire ecosystem, which would be a little bit too long and certainly useless.
I would like to select a few key players first, the largest ones, the ones which are described already already, apple, Microsoft and so on and so forth, but also a few additional key players in industry.
These companies are going to definitely play a role in the future including open AI for Chad, GPT, but some others and now will take the financial perspective.
Now observing the financial metrics of all these companies will lead us to a conclusion we have to go a little bit in theory of corporate finance.
It's not going to be a theoretical firm, but you will understand that a a little bit of refurbishment of the portfolio theory is probably a bit useful.
It's not going to be a debate on the threats and ethical issues which are created by the dramatic evolution of artificial intelligence.
For a few reasons.
First, it would take too long.
Second, it's not my purpose.
I'm not legitimate to deep dive into the insight of all these issues.
So we are going to concentrate on financial metrics.
Now, a few words of course about Chad gt.
Chad GT number three is offered to people in 2022, a year later end of 2023 chat, GPT-4, durable plus and so on so forth, and that's going to be amazing.
The propagation is going to be absolutely outstanding.
How much time does it take to take the first 1 million users Instagram, two and a half months? Wow, Netflix, three and a half years.
Cha five days.
The first 100 million users Instagram two and a half years Netflix, 10 euro cha, two months.
So it's absolutely unique in the history of high tech.
The speed of propagation of this fantastic application, which is T.
Now about market values on the NASDAQ of all these AI firms, you can take the six largest capitalization.
They are all involved more or less in the artificial intelligence business.
Six capitalization out of more than 3000 companies listed in the nasdaq, it's 0.2% of the number of firms.
The total capitalization of NASDAQ is more than 26,000 billions of dollars trillion dollars.
Now, these six largest capitalized agents, they represent around 17 trillion, which is 63%.
Now, if you, an extremely negligible number of firms represent about two thirds of the value, you understand that we are no more in a normal distribution in a Ian world, extremes have replaced the normal distribution.
Now, if you look at the theory of corporate finance, which is named the portfolio theory, when you look at the models thanks to which you can make some option valuation, blackshaw and so on so forth, it's based about log normal distribution of returns in the portfolio theory.
Now it's wrong.
It does not work anymore.
It's zero multiplied by infinity.
So the portfolio theory needs a little bit of refurbishment or maybe complete reconstruction.
Who has his sixth largest capitalization on nasdaq? The first one up to January 27 is Nvidia.
Nvidia generates revenues of about $100 billion multiplied by two each and every year.
By the way, the EBITDA is more than 60 billion, so you understand that the return on sales is absolutely outstanding, but free cash flow is a big diners.
Capital expenditures CapEx a billion.
Why? Because they don't manufacture the market.
Capitalization is stratospheric.
The price to book market value of equity divided by the book equity, which you are in, and the balance sheet is more than 60 and the enterprise value represents about 60 euro of EBDA, which is absolutely outstanding.
Interestingly, apple is number two.
The EBITDA is larger, 130 something.
The revenues of the last 12 months is about 400 billion, and capital expenditures are not that big because everything is outsourced at Apple, the market cap is not that far and the price to book is about the same.
Now, the enterprise value in EBITDA is 25, which is much less than 57 or 60 for Nvidia, and what is interesting is to observe the next four Microsoft, Google, Amazon and meta.
The revenues are from 150, 160 for meta to 620 billion for Amazon, which is getting closer and closer to Walmart.
By the way.
What is interesting is that the EBIT D rate is quite high, but for these companies from rank three to rank six, the capital expenditures are significantly higher because they invest in not only warehousing for Amazon but also in data centers, networks and so on so forth.
As a consequence, what is interesting to observe is that the price to book is a bit less because this company is in much more equity to finance their growth, to finance their capital expenditures, but the price to book is less and the enterprise value over EBITDA is quote normal, it's about 20 or a bit less than 20.
We all know companies which are not in the artificial intelligence world and whose enterprise value represents about 20 years of abit.
Why? Because these companies are profitable, they are prospects for growth and so on, so forth.
Of course, these companies are much smaller than the big companies, which I describe here, but the multiple is not completely stupid.
The last point I would like to discuss with you is the beta, the systematic risk coefficient.
It's quite high for Nvidia and what happened recently will probably make the beta go up again, but if you look at the other five, the beta is about to one.
You know what? These companies are completely embedded in economic life and this is why the beta is quite correlated with macroeconomic conditions about one.
Interestingly, if you make the parallel with the telecommunication business telecommunication equipment, the beta is much higher than telecommunication operators.
It's quite the same story, but I would like that you remember that the enterprise value over EBITDA of 20 is not abnormal.
A couple of conclusions about observing these figures.
Huge operating margins, ebitda, return on sales, absolutely fantastic, very significant free cash flows.
Sometimes the free cash flow is very close to the EBITDA because there's no CapEx.
Sometimes it's much less than the ebitda, but it's still quite big.
So you have very large companies with fantastic operating margins and strong free cash flows.
It's a fantastic and concentrated financial power.
A few additional comments about returning the cash to shareholders.
The return on equity for Apple is 150, 160 NVD is more than 100%.
The return on equity is the rate of return for the shareholders.
Now, if the return on equity is very high, it's quite normal that the price to book is very high and for these two companies, a price to book is exceeding 60, but the return on equity is not only the performance of the company, it is the ability to grow.
You remember this concept of sustainable growth rates, the ability of the company to self-finance its growth and then the ability to grow is more than 100% if you don't take into account returning the cash to shareholders.
Now, let's take these two companies one after the other.
Apple revenues are owe by about 10% per year earning soft tax 2024 $94 billion.
Just to give you an idea about transforming that into cash, the company generates a free cash flow of $1 million every five minutes.
So when you are profitable, when you are immensely profitable in generating a lot of cash and the gross rate is on lead 10%, you generate much more cash and what you need to final your growth, then you can return the cash to shareholders buy back 95 billion dividends, 15 billion, so we can return $110 billion.
But what is interesting about Nvidia, Nvidia is doubling the revenues each and every year.
So the revenue growth is 100% earnings after tax 2023 last, last 12 months and about 30 billion.
But as a company does not invest CapEx 1 billion, the company can return the cash to the shareholders dividends and buybacks altogether during the last two years, they represent $20 billion.
So you understand that these companies are extremely strong and they can give the money back to the shareholders.
What are shareholders going to do with this money? Reinvest in artificial intelligence.
So you have plenty your financial resources ready to invest in the development of this industry and generated by the leaders of this industry.
That's a huge value, but sometimes the value is fraud.
We can take the example of what happened to Google barred the artificial intelligence application of Google in February, 2023.
The company in the middle of a presentation is providing an inaccurate information.
Rer is in a room and immediately circulates a mistake.
What immediately happens in the stock market time magazine title an error just cost Google $100 billion.
So this figure is absolutely stratospheric, but it's extremely temporary.
If you look at the graph, the evolution of alphabet of Google throughout the years, you understand that it's up, it's down and there's a blip somewhere on the screen just for a few days, a few weeks, and then it's back.
It's back up.
The trend is there.
If you remember Facebook now meta with Cambridge Analytica.
Cambridge Analytica was very much an issue for Facebook.
The third price went down.
Three weeks later it was up and it started going up again.
So it's just an accident.
It's a $100 billion temporary accident.
Now, AI is not limited to the big six, which I mentioned.
Open AI of course is in the middle of the picture and tropic x ai, mytral ai, a few mobility cases and Reddit case.
It's what I'm going to discuss now with you.
Let's start with open ai.
OpenAI, very well known Chad g PT founded in 2015 by Sam Altman, who a very good friend of Sam Alman, who's name is Elon Musk and a few others.
It seems that Sam Atman and Elon Musk are no more friends anymore, which happen in uh, in number of cases in this industry.
Why did they create open AI for the benefit of humankind? The free cash flow is never that far from that.
Now, at that time, Sam Atman is the president of an outstanding accelerator based in California.
His name is Y Combinator.
In 2019, he leaves Y Combinator to be 100% CEO of open ai.
He's going to be host stood by the board and then back in the board, right? And then a few people are going to be out in the board and top managers, top people at OpenAI, three co-founders, C-S-O-C-T-O security r and d manager and a number of people are living open ai, which is definitely an issue for the company in the future as far as some money is concerned is still there and what's happened to the value of the company? Well, in 2023, Microsoft, the long term financial and business partner and technological partner of open AI provided $10 billion.
This amount has to be compared with the primo valuation.
So the value before you receive the cash of 19 billion.
So post money, the value is 19%, which is 29, about 30 billion a year later, exactly 18 months later.
In 2024, an additional cash injection of $6.6 billion out of a premium of 150.
Now the 150 has to be compared with 30, and so the value of the company was multiplied by five in just 18 months.
So we are not exactly in the same valuation process as normal companies and normal industries.
Tropic is a company which was founded by former open AI people because a number of them left the company.
The company was founded in 2021.
The last valuation of the company series D, February, 2024 mentions about 17 to $18 billion.
A brother and a sister created the company.
It seems that they signed a big contract with Amazon to provide service and so on so forth, and they received some financing interestingly from Google, from the speculative fund, which was created by SoftBank Vision Fund.
Keep in mind, SoftBank, we're going to talk about the capital later and the financing was about $4 billion.
The valuation are stratospheric, but it's just paper valuation.
The MONA is available in cash.
Elon Musk is no more friend of uh, Sam man.
He left, uh, open ai.
He said, we have to stop everything.
Please stop everything because in the meantime, I'm creating my own company.
His name is XAIX AI because of SpaceX, because of Twitter becoming X and so on.
The company is founded in July, 2023.
The name of the application is extremely poetic gr and in May, 2024, the company leads Monet $6 billion out of a pre month valuation of 18 billion who participated to the equity issue, the usual suspect, Sequoia plus a few sovereign funds and so on so forth.
It was in May Primo 18 in December, we need an additional $6 billion, PRI Monet, 44 billion.
So again, the value is up, but the value is not cash.
What is injected in these companies to pay the salaries is cash.
In Europe.
In France, there's a company which seems to be quite promising in the same industry name is Mistrial ai.
We have a contract with Microsoft and they have a nice credibility in terms of technology.
The company was created in 20 23, 3 founders coming from Google DeepMind and from Meta Facebook, but in order to raise just $1.1 billion, the company needed 40 investors, which is complete different context in terms of venture capital series B in June, 2024.
Now, premium evaluation is 5.2 billion.
The capital is raising half a billion.
The name of the application is Lucia Lecha, so it's about chat, but Lucia is a cartoon humoristic cartoon in Europe, which is extremely well known and very much appreciated by the way.
Now if you look at what Crunchbase very serious website is giving us a list of who received more than $1 billion in the last two years.
23 companies, of course the usual suspects I already mentioned, but two companies in the mobility business and it's interesting to spend a minute on that.
The first one is Waymo Ymo is Google mobility activity and the company received in October 20, 24, 5 $0.6 billion to develop autonomous driving and so on so forth.
Cruise is also a very interesting case.
It's about self-driving vehicles, it's about robo taxis.
The company received plenty of money from GM and SoftBank Vision Fund, but now GM is a controlling interest in growth.
I will describe a bit later.
The company received 2.2 in 2022 and 850 million in 2024 provided by General Motor.
It is a total of $16 billion of funds received from investors, including the largest one general models.
Now, again, the value is fragile.
GM grew, it's not a great investment.
They developed a business fleet of Robax.
The company took off in 2018 with GM and SoftBank Vision Fund.
SoftBank provided $2 billion and GM bought the Southbank Vision fund stocks in 2022, and now the company has almost the whole control, the entire control of uh, CRO revenues generated by the company throughout the years.
It's about half a billion, but GM invested $10 billion and it's not successful at all.
This is why GM recently said, we're going to stop everything about that.
We are going to stop the robo taxi business.
We are going to take the technology and bring the technology back to GM for personal vehicles, driving assistance and so on so forth.
And instead of costing 10 billion wasted money, we are going to invest about a billion per year.
So GM had bought the SoftBank Vision Fund shares, they had 90%.
Now they have an option to go to 97%, they're going to buy back all the share end of the story, and Cruise is not a success.
So it's not only about successes in this industry and sometimes life stops.
Reddit is the last case I would like to discuss with you.
You can't find on the academy an educational film dated April, 2024, not that far ago, and it was about E-S-G-E-S-G, which potentially creates value thanks to the evolution of ESG.
At Reddit, the company could be listed and so it created value for its shareholders.
Reddit is a community of communities, a sub Reddit company created in 2005.
In 2014, series B, the company is raising $50 million.
Most of the money comes from Sam Atman and was a good investment for Mr.
Atman because the value of his equity stake in Reddi is supposedly about $600 million.
Today, thanks to becoming ESG friendly, Reddit became public company could go for an IPO.
The IPO started at $8 billion.
Now it's about $11 billion.
But the question is what is the main asset of Reddi? Reddi is publishing is communicating on its strategy and the evolution of the business model.
First, it's about advertising, segmentation, positioning, data return investment for those people who need advertising.
Great second layer data licensing, all these people discussing, interacting in the communities and in the subreddits.
It's a fantastic playground to train your AI applications.
And so at the end of the day, Reddit is making money selling to big companies the usual suspects.
I already imagine selling the access to the subreddits so that they can train the application and you know how much it costs.
You know how strategy and critical this activity is for those people, those companies providing the AI applications.
Now, I provided some information about large and less, large, but key players in this industry.
There are plenty of remaining issues.
I said I'm not going to discuss threats on ethics, but it's nice to mention that these companies are consuming a lot of energy.
They're issuing plenty of carbon.
It's a oligo poly, an extremely restricted oligo poly.
Not many companies with fantastic resources.
Valuation and cash sovereignty is a little bit of problem consumption of metals and related with that, the recycling issue, the data centers, they consume huge space, which is not available to do anything else.
These companies, they have plenty of cash and they attract competencies and technical competencies are quite scarce, but as they have plenty of money, they can offer extremely attractive remuneration for these people.
This competence is now working for artificial intelligence, are not available for the classical and traditional industries.
This is the advantage of being strong, powerful, and full of financial resources.
There are also plenty of issues about cyber security, et cetera, et cetera.
A couple of very recent examples to conclude this film.
You remember in the nineties, Stargate Stargate was a science fiction film.
Now 20, 25, January the 22nd of January, what is announced by open AI here on X, we have created the Stargate project.
It's going to be joint venture with a few actors including open ai, of course, including SoftBank, Mai or Csan, including Larry Ellison Rockle, and we are going to invest $500 billion in artificial intelligence over the next four years.
That's absolutely outstanding.
I already did discussed Masayoshi son investment cycle at SoftBank, SoftBank Vision Fund and so on.
Now, the person is very much involved in ai.
The opportunity I had to discuss about some investments where about we work, which was not exactly successful, but Massa Oran says we have to bring brain computer to life.
So we have to be able to transform the biological brain into a set of algorithm combined with microchips.
Those this is brain, computer to life.
This is a fantastic dream.
Successful or not, we are going to see that in the future.
But what is interesting that SoftBank was in a number of different businesses and now they are a little bit catching up about ai.
They had a very good start buying ARM holdings, the company, which Nvidia wanted to buy by the way, and the antitrust author said, don't even think about it when they bought arm, it is in 2060 and they pay $31 billion for the acquisition.
2023, the company goes public at the value, which is 54 billion.
Billion.
That's a good deal, and today it's 155.
That's a good investment.
Now, the very best investment ever made by Mr.
Masai Oran in the tech business is Alibaba.
He invested $100 million, which was transformed eventually into tens of billions of dollars.
The second one is Nvidia.
Congratulations because he invested $4 billion in May, 2017 when the market cap was about $80 billion.
It was already a quite high figure, but it has been modified by 40 since then.
He invested in cruise.
You remember the GM story? He invested two point something billion in 2018 and sold in 2022 for two point billion something plus minus zero per cm open AI yet, but quite late in October, 2024.
So NVDI is definitely a very high success arm, is quite good profitability.
And so SoftBank Vision Fund is a very significant actor in ai.
This is why, and you see that on the picture.
The president of the United States is welcoming maan together with Larry Ellison and Samba.
The joint venture is going to be funded by OpenAI, 19 billion, soft bank, 19 billion or rl.
We don't know exactly how much plus one or two additional, and at the end of the day it's going to be $500 billion over the next four years.
But on this picture, there are two people who are absolutely missing.
The first one is Microsoft.
Microsoft originally was a financial partner.
They provide a lot of funding for open ai.
Now they say, oh, this is going to be a very strong technology partner together with Arm, SoftBank, Microsoft, Nvidia, or Rackl, et cetera, et cetera.
Microsoft is no more a financial partner.
Maybe it's a technological called partner.
So it's now tier two in the process.
The other one who is missing is Elon Musk.
Elon Musk co-created open AI with Sam Atman.
Now they are not exactly friends and Elon Musk brought open AI and Sam Atman to the court now adding Microsoft antitrust claims.
We are November, 2024 and definitely Mr.
Maersk, who seems to be good friend with the president of the United States is not a good friend of Stargate Project.
The second event, which extremely recently happened, January 27, is deep seek a Chinese startup and they say, okay, we are doing good job.
We are proposing an application which is almost as good as Chad G, but we didn't pass same price as Chad G.
In fact, developing the application cost about 3% of what it cost to open AI to develop Chad, GPT 1, 2, 3, 4, et cetera, et cetera.
And the consequence, what happened to Nvidia stock price, it went down dramatically.
The company lost $600 billion of market cap in one day, $1 trillion evaporated in one day as a consequence of Dipe announcement and what the market observe about the efficiency of the application.
Why? Because very likely all these investments are probably overestimated in terms of value creation.
And there are a few guys here over there who can do a quite good job spending much less than all these big companies, which I already mentioned.
Consequence Nvidia Verage down together with some other companies.
Now the Chinese AI business looks a little bit better, and this is why, for example, a company whose name is Merit Interactive.
So the stock price is skyrocketing because the one man's misfortune is outta the man's gain in finance with him that the portfolio and diversifying the specific risk.
So if these guys are able to do a good job spending much less, the value is much higher stock cross.
He's plus 20% in one day, which is absolutely outstanding and drives me to the first of my four conclusions.
The first one is when you watch this video, unfortunately it's going to be obsolete because life moves on very quickly and from the moment I recorded the video and the moment you watched it, plenty of things will have happened, which makes this video a little bit obsolete.
At least it'll give you some information about the past hopefully second conclusion, which is a little bit more serious.
The financial power is hugely concentrated in an extremely limited number of hands.
I described the six big players plus a few additional.
You understand their financial power.
It's a oligopoly, it's a strong oligopoly with outstanding operating margins, huge free cash flows, fantastic power to develop the business and take the power and take the control.
Hugely concentrated power, certainly not goals curve.
And now we have to reconsider, we have to rebuild the corporate finance theory.
It's no more, it's normal, a log normal distribution.
It is zero multiplied by infinity.
It's completely different context.
We have to start from scratch again and we have observed friends Elon Musk with Sam Atman, Microsoft yesterday.
Microsoft Today.
And about friendship, it reminds me is a very well known Raider on Wall Street, Carl Aiken, who used to say about friendship.
If you want a friend, you should buy yourself a dog.
It's about friendship in artificial intelligence, in finance, and on Wall Street.
Thank you very much.