1
00:00:01,920 --> 00:00:04,520
Hey everybody. 
Welcome back to the Elon Musk 

2
00:00:04,520 --> 00:00:07,120
Podcast. 
This is a show where we discuss 

3
00:00:07,320 --> 00:00:11,520
the critical crossroads, the 
Shape, SpaceX, Tesla X, The 

4
00:00:11,520 --> 00:00:15,600
Boring Company, and Neurolink. 
I'm your host Will Walden. 

5
00:00:17,800 --> 00:00:21,880
Sam Altman is working on a brain
chip venture that may try 

6
00:00:21,880 --> 00:00:26,200
something no major competitor 
has attempted before, changing 

7
00:00:26,200 --> 00:00:30,080
how brain cells work using gene 
therapy. 

8
00:00:30,640 --> 00:00:34,520
Merge Labs, which is the startup
associated with Altman and Open 

9
00:00:34,560 --> 00:00:38,200
AI's venture arm, is considering
a method that would genetically 

10
00:00:38,200 --> 00:00:41,480
alter brain cells so they can 
better interface with implanted 

11
00:00:41,480 --> 00:00:44,600
technology, People familiar with
the plans say. 

12
00:00:44,600 --> 00:00:48,320
This includes using ultrasound 
to communicate with those 

13
00:00:48,320 --> 00:00:51,400
altered cells. 
Now the goal is to build a brain

14
00:00:51,400 --> 00:00:54,360
computer interface that allows 
seamless interaction between 

15
00:00:54,360 --> 00:01:00,400
human thought and AI tools like 
ChatGPT, Altman said publicly. 

16
00:01:00,400 --> 00:01:03,200
He wants to be able to think 
something and get a ChatGPT 

17
00:01:03,200 --> 00:01:07,040
response without needing to 
speak or type, you know? 

18
00:01:07,040 --> 00:01:10,160
This idea sits far outside what 
companies like Neurolink 

19
00:01:10,320 --> 00:01:13,080
currently focus on. 
Elon Musk's firm is pushing 

20
00:01:13,080 --> 00:01:17,000
forward with electrode based 
implants that read and stimulate

21
00:01:17,000 --> 00:01:21,120
brain activity electronically. 
And Neurolink raised $650 

22
00:01:21,120 --> 00:01:26,520
million this year and now holds 
a valuation of $9 billion. 

23
00:01:27,000 --> 00:01:29,120
Merge Labs wants to go a 
different direction, though, 

24
00:01:29,440 --> 00:01:33,600
tapping into sonogenetics, which
is a research field that studies

25
00:01:33,600 --> 00:01:36,640
how to make cells respond to 
ultrasound through genetic 

26
00:01:36,640 --> 00:01:39,640
engineering. 
Sonogenetics isn't a new field, 

27
00:01:39,920 --> 00:01:43,920
but using it in a commercial 
brain chip product would be a 

28
00:01:43,920 --> 00:01:46,840
major leap forward. 
Researchers have spent years 

29
00:01:46,840 --> 00:01:50,000
exploring how cells can be 
sensitized to sound waves. 

30
00:01:50,440 --> 00:01:54,080
Merge is evaluating whether gene
therapy can make specific 

31
00:01:54,080 --> 00:01:57,760
neurons react predictably to 
ultrasound, which could then be 

32
00:01:57,760 --> 00:02:00,200
used to send and receive brain 
signals. 

33
00:02:00,680 --> 00:02:03,800
This method may allow less 
invasive implants, since 

34
00:02:03,800 --> 00:02:07,000
ultrasound can pass through 
tissues more easily than 

35
00:02:07,040 --> 00:02:10,759
electrical signals. 
The hardware involved would also

36
00:02:10,759 --> 00:02:12,800
break new ground. 
Sources say that Merge is 

37
00:02:12,800 --> 00:02:15,600
looking at embedding an 
ultrasound device into the human

38
00:02:15,600 --> 00:02:18,040
skull to work with the modified 
brain tissue. 

39
00:02:18,440 --> 00:02:21,920
This setup could allow for real 
time detection and control of 

40
00:02:21,920 --> 00:02:25,560
neurological activity without 
relying on wires or probes 

41
00:02:25,560 --> 00:02:27,880
inserted deep inside of your 
brain. 

42
00:02:28,360 --> 00:02:32,040
And if this technology ends up 
working, it may enable smoother 

43
00:02:32,040 --> 00:02:35,160
interaction between the brain 
and digital systems, with fewer 

44
00:02:35,160 --> 00:02:37,480
physical risks than traditional 
electrodes. 

45
00:02:38,240 --> 00:02:41,240
Now, Open AI has not confirmed 
the details publicly, though, 

46
00:02:41,440 --> 00:02:44,480
and Altman said the deal has not
been finalized. 

47
00:02:44,640 --> 00:02:48,480
Still, Altman's presence around 
the project is very visible. 

48
00:02:48,760 --> 00:02:52,280
He plans to Co found the 
company, though he will not take

49
00:02:52,280 --> 00:02:54,720
an operational role or 
personally invest in it. 

50
00:02:55,120 --> 00:02:58,680
Much of the early funding, 
expected a total $250 million, 

51
00:02:58,680 --> 00:03:00,240
will come from Open a Eyes 
venture. 

52
00:03:00,240 --> 00:03:04,160
Arm Merge is currently valued at
850 million, according to three 

53
00:03:04,160 --> 00:03:07,440
people with direct knowledge of 
the funding talks. 

54
00:03:08,760 --> 00:03:13,360
Altman is working on the project
alongside Alex Blania, who 

55
00:03:13,360 --> 00:03:17,120
currently leads Worldcoin, which
is a separate venture that uses 

56
00:03:17,160 --> 00:03:20,680
iris scans to verify digital 
identities. 

57
00:03:21,080 --> 00:03:24,960
Blania has close ties to Almond 
through Worldcoin, another 

58
00:03:24,960 --> 00:03:26,760
project backed by Opening Eye 
leadership. 

59
00:03:27,160 --> 00:03:29,800
Altman previously invested in 
Musk's Neuralink, but the two 

60
00:03:29,800 --> 00:03:33,240
entrepreneurs are now not quite 
friends, and they're now 

61
00:03:33,240 --> 00:03:35,840
competing head to head in the 
brain interface space. 

62
00:03:36,400 --> 00:03:40,800
The rivalry escalated after Musk
sued Open AI for straying from 

63
00:03:40,800 --> 00:03:44,440
its original nonprofit mission. 
Now brain computer interface, or

64
00:03:44,440 --> 00:03:48,200
BCI, companies are multiplying, 
but few are pursuing the path 

65
00:03:48,200 --> 00:03:51,440
Merge appears to be taking. 
Nerlink is still the most high 

66
00:03:51,440 --> 00:03:55,040
profile BCI company and has 
begun human testing for treating

67
00:03:55,040 --> 00:03:57,240
paralysis. 
Others like Precision 

68
00:03:57,240 --> 00:04:01,640
Neuroscience and Synchron focus 
on non invasive or semi invasive

69
00:04:01,640 --> 00:04:04,400
implants designed to assist 
people with neurological 

70
00:04:04,400 --> 00:04:06,720
disorders. 
These companies rely on reading 

71
00:04:06,720 --> 00:04:09,880
electrical signals from the 
brain, often requiring surgical 

72
00:04:09,880 --> 00:04:14,040
insertion of electrodes. 
Emerge stands out by potentially

73
00:04:14,280 --> 00:04:17,279
combining genetic engineering 
with ultrasound, aiming for a 

74
00:04:17,279 --> 00:04:20,440
system that works through 
acoustic stimulation of modified

75
00:04:20,440 --> 00:04:23,560
brain tissue. 
That route could avoid problems 

76
00:04:23,560 --> 00:04:26,600
associated with electrodes like 
scar tissue or long term signal 

77
00:04:26,600 --> 00:04:29,000
degradation. 
However, it could also take 

78
00:04:29,200 --> 00:04:32,040
years before the safety and 
reliability of this approach is 

79
00:04:32,040 --> 00:04:34,920
proven in humans, and the 
technology will require 

80
00:04:35,280 --> 00:04:39,040
overcoming major scientific 
hurdles, particularly around 

81
00:04:39,160 --> 00:04:42,040
targeted gene delivery and long 
term tissue response. 

82
00:04:42,640 --> 00:04:45,920
Ultrasound has gained momentum 
in neuroscience research as a 

83
00:04:45,960 --> 00:04:48,400
non evasive way to influence 
brain activity. 

84
00:04:48,800 --> 00:04:52,280
Start-ups like Nudge in Others, 
backed by investors including 

85
00:04:52,560 --> 00:04:56,720
Fred ESM and Reed Hoffman, are 
developing helmets that are 

86
00:04:56,720 --> 00:05:00,560
focused on using ultrasound to 
stimulate specific brain 

87
00:05:00,560 --> 00:05:03,000
regions. 
These devices are aimed at 

88
00:05:03,000 --> 00:05:06,880
treating psychiatric disorders 
without implants, and Merge 

89
00:05:06,880 --> 00:05:09,560
would be among the first to 
pursue implanted ultrasound 

90
00:05:09,840 --> 00:05:15,600
combined with cell modification.
Now, Open AI has not officially 

91
00:05:15,600 --> 00:05:18,280
commented on its role in the 
Merge product. 

92
00:05:18,640 --> 00:05:21,600
However, the company's venture 
fund is expected to provide a 

93
00:05:21,680 --> 00:05:24,320
large share of the capital 
needed to bring the project to 

94
00:05:24,320 --> 00:05:26,800
life. 
And if successful, Merge could 

95
00:05:26,800 --> 00:05:30,800
redefine how humans interact 
with AI systems, bypassing 

96
00:05:30,800 --> 00:05:35,080
screens and keyboards for 
something closer to direct 

97
00:05:35,080 --> 00:05:38,080
thought transmission. 
And it's still an early 

98
00:05:38,080 --> 00:05:41,160
development, and the gene 
therapy approach could shift or 

99
00:05:41,160 --> 00:05:43,800
be scrapped, depending on 
technical feasibility. 

100
00:05:44,240 --> 00:05:46,840
And But for now, Altman is 
backing a plan that challenges 

101
00:05:46,840 --> 00:05:51,280
the standard brain chip road map
by blending genetics, sound, and

102
00:05:51,280 --> 00:05:53,800
computing. 
Hey, thank you so much for 

103
00:05:53,800 --> 00:05:55,960
listening today. 
I really do appreciate your 

104
00:05:55,960 --> 00:05:57,520
support. 
If you could take a second and 

105
00:05:57,520 --> 00:06:00,480
hit the subscribe or the follow 
button on whatever podcast 

106
00:06:00,480 --> 00:06:03,480
platform that you're listening 
on right now, I greatly 

107
00:06:03,480 --> 00:06:05,120
appreciate it. 
It helps out the show 

108
00:06:05,120 --> 00:06:07,760
tremendously, and you'll never 
miss an episode. 

109
00:06:08,000 --> 00:06:11,360
And each episode is about 10 
minutes or less to get you 

110
00:06:11,360 --> 00:06:14,080
caught up quickly. 
And please, if you want to 

111
00:06:14,080 --> 00:06:20,200
support the show even more, go 
to patreon.com/stagezero and 

112
00:06:20,200 --> 00:06:22,640
please take care of yourselves 
and each other and I'll see you 

113
00:06:22,640 --> 00:06:23,160
tomorrow.