aboutsummaryrefslogtreecommitdiff
path: root/paper.ms
blob: 863057974efdc477a0194e4cd6464d8be4154df8 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
.R1
short-label D.y
sort
.R2
.TL
Cryptography, crime, terror, and surveillance
.AU
\f[TI]Mohit Agarwal
.AI
February 2022
.LP
Modern encryption methods allow a level of privacy in communication
that has not before been seen: information that is encrypted cannot be
decrypted without the necessary keys, which in the case of RSA is
ensured by the large primes involved and the current intractability of
large prime factorisation. This allows for communication that is
practically guaranteed to be private: a relatively new phenomenon in
communications, seen with inventions such as the one-tme pad (cite)
which was cryptographically secure and used by the both the KGB and
NSA (cite), beyond the use of the Enigma and Lorentz machines by the
Nazis which were both of which were decrypted by cryptanalysis methods
during the Second World War. Today, secure cryptographic methods are
used not only by government backed agencies in preventing or
practising espionage, but by individual citizens who are interested in
their privacy, security, or are simply using a program that happens to
encrypt their communications. Naturally, current availability of
cryptography potentially allows for malicious actors such as criminals
or terrorists to use encryption in order to commit crimes or acts of
terror.
In
response to the threats of encryption and communications technology
generally, governments have often engaged in signals intelligence
(SIGINT) such as phone line tapping. Modern SIGINT initiatives have
become incredibly complex and sophisticated and have grown greatly as
popular adoption of technology has grown. Part of government interest
in SIGINT is a direct response to perceived threads, such as the
PATRIOT Act in the US which followed the 2001 terrorist attacks with
the objective of strengthening national security (cite). Later, the
FISA Amendments Act of 2008  further increased increased the powers of
law enforcement to access information, such as allowing the Attorney
General and Director of National Intelligence to provide information
about individuals outside the United States {House bill FISA}. It was,
however, the PATRIOT Act and FISA Amendments Act that was the
justification for large scale surveillance including the records of
phone calls of customers of the Verizon network, including calls from
the US to other states as well as calls localised entirely within the
US {guardian greenwald verizon}{guardian NSA roberts}{times savage
2013}. State sponsored SIGINT programmes such as that in the US aims
to respond to encryption and other technological developments with the
primary interest of overcoming it in order to prevent terror and
crime. These measures have, however, had arguably limited
effectiveness and have violated the privacy of individuals who are
not suspected of being a threat to national security. Responses to
encryption domestically and internationally will have significant
consequences, given the potential importance of the information being
communicated. Successful SIGINT and cryptanalysis by government
agencies can successfully respond to modern threats of crime and
terror. A failure of responsible governance, however may not only
threaten the privacy of individuals unnecessarily, but also fail to
respond to the ways in which criminals and terrorists are using
encryption existing thereby only as a tool of authoritarian control.

An argument is often made against allowing widespread use of
encryption and generally against widespread effective operations
security (OPSEC) in the public sector in the interest of
national security. With access to communications and usage history law
enforcement and government can quickly discover large amounts of
information useful in a criminal investigation or in preventing
criminal activity. Graham{#CTC terrorists} explores the use of
encryption by terrorists which is often cited in a reason for giving
governments access to unencrypted Internet communications so that
suspicious activity can be flagged and investigated in order to
prevent a terror attack or in order to better respond in the case of
an attack. Graham describes the extensive use of end to end encryption
used by terrorists in order to avoid interception by the authorities.
Due to U.S. usage of intercepted communications to uncover and prevent a
number of al-Qa'ida plots, the terrorist organisation and other
terrorist groups have increasingly used encrypted communications (read
citation from Graham). An significant factor is the use of
non-mainstreams software in early use of encryption by terrorists,
including a program that built a wrapper around the popular, secure,
and open source PGP called \fIMujahedeen secrets\fR. Although now
terrorists and criminals use widely available, popular, and
user-friendly software such as the Tails operating system or Telegram
(Graham citation 28), terrorists organisations have shown an ability
to make use of more obscure and complicated systems, as well as use
publicly available source code in order to construct software for
operatives to use.

Although the issue of popular messaging technologies and their support
for 'end-to-end encryption' is often discussed, the argument that the
introduction of end-to-end encryption by large companies such as
Facebook gives an advantage to criminals {conversation Facebook}{home
office} is arguably an entirely invalid one. By preventing the usage
of true end-to-end encryption in industry, we will not be able to
prevent those attempting to evade the law from doing so, as shown in
the case of terrorist organisations who have used more obscure
software in the past and also in the case of the abundance of illegal
activity that occurs on the so called dark web in the form of the
trade of drugs and child pornography among others (cite). Instead the
limitation of use of encryption on popular software will only decrease
the privacy of those uninterested in criminal activity and instead
using technology to communicate. In the case of platforms such as
Instagram (which is owned by Facebook) it is quite clear that the vast
majority of communications (cite) will not contain anything illegal
(reword) and that it is these conversations that will suffer from a
lack of encryption. The information exposed by Edward Snowden in 2013
demonstrates that the US government has processed and collected vast
amounts of unencrypted data (cite) and likely continues to do so. In
the case of unencrypted messaging the problem remains and preventing
end to end encryption will simply allow governments to maintain the
status quo of being able to intercept and read all communications
between its citizens and individuals outside of their jurisdictions.

In order to conduct the vast amounts of surveillance they did in the
GDR (German Democratic Republic) in support of the ruling party
{Jarausch}, the Stasi gathered information from a vast network of
informants who greatly outnumbered Stasi agents {Bruce 2014}. Whilst
in Nazi Germany there may have been around one Gestapo agent for every
2300 citizens, in the GDR it was closer to one informant or officer for
every 63 citizens. Those living in the GDR often had experiences
involving investigation by the Stasi and there was clearly an
understanding amongst citizens {funder} of the GDR that one had to be
wary of an informant or agent listening in. In modern western society
there is a similar collective understanding that governments
attempting to carry out surveillance on a massive scale on their own
citizens. A key distinction, however, is that in societies such as the
UK, this work is not carried out by a vast network of informants,
there are no gargantuan gargantuan stores of paper, and there are no
hundreds of miles of film (cite all) documenting and aiding the
surveillance of the authorities. Instead, the level of surveillance
that large, secretive groups of individuals once had to carry out in
order to enable a surveillance state can be performed instead through
bureaucracies and technological methods. In modern times, governments
can operate with a very limited number of operatives `on the ground`,
and instead focus attention on the giant amounts of data they have for
processing in order to make the findings they intend to: be it crime,
terrorism, or - as was the case with the Gestapo and Stasi - descent.

As with any technology, regulation has followed behind development in
an attempt to control its limits. Much as automotive regulation
followed the increase in popularity of cars in areas such as the UK
and US, regulation will no doubt follow the newfound popularity of
heavy encryption. There are however, difference in the case of
encryption when compared to cars. The rate of change with modern
technology is far greater. 
In the case of encryption regulation will continuously struggle to
control encryption methods due in part to how quickly they change, but
perhaps moreso due to their decentralised nature, where a government
cannot prevent the existence of software that enables encryption which
is open source and reproducible internationally. Just as media privacy
through torrents and access to hidden services over tor are possible
without significant regulation, regulation of encryption may prove
impossible. An arguably useful tool to the authorities does exist in
the hardware and infrastructure that users of the internet rely on.
Firstly, the vast majority (cite) of users in the foreseeable future
will continue to use the highly popular CPUs designed by Intel. 
Concerns have already been expressed {Intel Management portnoy} with regard to
the Intel Management Engine that exists on modern processors produced
by Intel. Should governments chose that backdoor access is essential,
then this presence in hardware around the world alongside an influence
over Intel (a US based company) to give access to governments may
provide them with the ability to access information directly from the
target's hardware rather than having to intercept information in
transit. This would go for other hardware vendors such as AMD or ARM
also. Whether or not companies such as Intel would open backdoors to
governments is up for debate, however we are aware that in the case of
the Intel Management there was potentially an ability for it to be
disabled by US government authorities such as the NSA, demonstrating a
level of leverage the US government potentially has over organisations
including but not limited to Intel {register kill switch}{intel me
bleepingcomputer}.
Regardless of the level of influence governments might or
might not hold over private corporations, the potential exists for
systems built into non-open hardware which most people, even those
using open software use, leaving them more open to exploitation from
either state or private actors. Furthermore, there is a visible
interest in increasing the presence of technologies on the hardware
level, including the aforementioned Intel Management Engine, the
Trusted Platform Module (cite), and recently Microsoft's Pluton (cite)
subsystem, which will be present on hardware sold in the future. This
variety of hardware within a single computer is a rather interesting
and potentially worrying development, particularly with the clear
level influence, interest, and competitiveness both the US {US House
chip manufacturing bill} and Chinese governments (cite) are
respectively showing (the US and China are the two largest chip
manufacturers (cite, reword)). In light of potential issues with
hardware in a privacy sense, there have been developments in `open
hardware'. RISC V is an instruction set for processors, that, opposed
to ARM, Intel, and AMD which are developed in secret, RISC V is an open
standard originating from the University of California, Berkeley (UC
Berkeley). This therefore allows for open source CPU designs, such as
those designed at UC Berkeley, as well as those from other parties,
such as Alibaba Group (cite all). A significant amount of existing
software has been ported to the RISC V platform (cite) and been
implemented commercially by companies such as Google, for a security
module in the `Pixel 6' smartphone (cite). This attention and interest
in the technology potentially indicates a shift in attitude and want
for more open hardware and a general concern for the source of
computing equipment. Examples, such as a laptop created by the
manufacturer Frame Work Inc which aims to be more expandable,
serviceable and repairable then existing laptops, gaining significant
media coverage (cite) further show an interest from the public in open
hardware. An argument can be made that such projects are for niche
interest groups only, and that such solutions will never see the
commercial success seen by the larger, non-open manufacturers such as
Intel and ARM, however clear adoption of standards such as RISC V by
large institutions (cite) as well as the clear interest the public
have demonstrated in commercially available open solutions (research,
cite) demonstrate quite the opposite: that open hardware will continue
to become increasingly prevalent and that currently popular hardware
with its susceptibility to surveillance will possibly start to
disappear. 

A shift toward open standards reveals a problem for law enforcement
agencies and counterterrorism forces. The tools of mass surveillance
that once enabled investigation into crime or terror such as reading
messages/emails, listening to calls, tracking location, or analysing
metadata (cite?) may no longer be effective, thereby potentially
preventing such investigation to occur. For governments, this is
arguably the result of such heavy surveillance in the first place.
It is clear that knowledge such as the 2013 Snowden leaks had an impact
on the public (cite), and that people are therby more interested in
their privacy and preventing surveillance. The exception to this has
been in China, where the government has unparalleled control over the
flow of information over the internet. This has allowed the filtering
of content, prevention from accessing sites, and the blocking of the
anonymity network Tor which would allow users to circumvent measures
put in place by the government {firewall} (cite for Tor). (research:
would such measures even work in western world?)

In addition, the rate of development in unconventional computing
methods is increasing rapidly. Effective quantum computing will
mean that existing popular cryptographic algorithms such as RSA will
no longer be secure due to the potential for computations that would
take unreasonable amounts of time on classical computers to be solved
quickly (reword) such as prime factorisation on which RSA encryption
relies {lily chen quantum}. RSA encryption is currently in use for
applications such as private communications and digital signatures.
Significant research such as at IBM in recent years (cite) has shown
feasibility in current ideas surrounding quantum computing and
promising results in development towards quantum supremacy and in the
future the breakdown of current cryptographic methods.

Indeed, there
have already been claims (although disputed) to quantum supremacy from

Once more, the significant research is occurring as aforementioned in
the US and in China {quantum research in china}. Both in the US at
Google {google supremacy nature} and in China {china quantum
advantage}{science photons quantum advantage}. 

Is discussion on this useful?
Individuals around the world have clearly expressed interest in
matters of privacy and encryption (cite) and open source software
allows those with the technical skills to become involved in the
development of technology that enables strong encryption and avoids
state surveillance. Measures taken by governments to prevent this
development will doubtless be limited unless extreme actions such as
those seen in China are taken. Otherwise, development will continue to
occur in both free and non free societies in support of individual
freedoms. The assertion of `Linus' law` that "given enough eyeballs,
all bugs are shallow" (cite - CathBaz) creates a serious inability
for actors such as governments to engineer backdoors into software as
the NSA previously has (cite) or to prevent the development of
software altogether (find example). On the other hand, a significant
amount of the software and hardware 

The discussion of encryption and related technologies has arguably
limited impact. State actors such as the NSA will continue to act
against individual freedoms and attempt to find or introduce backdoors
in technology that is widely used as part of its actions purportedly
in the interest of `national security`. Although public reactions to
information such as the 2013 Edward Snowden releases have been very
strong, they have not had significant effects on legislature, the
funding received by the NSA, and quite possibly the level of
surveillance carried out by the NSA (cite all). Thus, from recent
history, discussions in public or private spheres are unlikely to
influence decisions made inside already secretive agencies where
governments are ready to except that sacrifices must be made for the
greater good. Of course, the issue arises when surveillance exists
that does not exist simply to protect a nation, but instead mass,
indiscriminate surveillance is carried out on citizens not suspected
of any criminal or terrorist activity such as the Optic Nerve
program in the United Kingdom (cite), however governments nonetheless
prove willing to fund the activities of surveillance agencies.
Furthermore, there are options available to authorities that are
regularly made use of. (Give example from Graham)

Modern cryptographic algorithms are `cryptographically secure`; the
underlying theoretical concepts mean that breaking the encryption to
intercept a communication is possible only through a brute-force
attack and is therefore, due to the nature of the algorithm. This
however, does not consider implementational flaws. Indeed,
implementational flaws are the ways in which modern breaks of
algorithms such as RSA (cite) occur, and methods such as timing
attacks (cite) and voltage level analysis attacks, as well as memory
attacks (cold boot, rubber hose ...) (do some light explaining) (cite
all) have the potential to overcome any level of sophistication that
cryptographic algorithms may have, and simply give away information
such as keys (research, cite).

.nr HY 0
.ad l
Intro
    Foreign Intelligence Surveillance Act of 1978 Amendments Act of 2008
	USA FREEDOM Act (2015)(HR 2048)

Cryptography
    https://wikiless.org/wiki/Kerckhoffs%27s_principle?lang=en
    Timing Attacks
        RSA

Spectre and Meltdown (disucss speculative execution)
    https://googleprojectzero.blogspot.com/2018/01/reading-privileged-memory-with-side.html
    https://www.nytimes.com/2018/01/03/business/computer-flaws.html
    https://support.apple.com/en-us/HT208394
    https://www.ibm.com/blogs/psirt/potential-cpu-security-issue/
    https://www.ibm.com/blogs/psirt/potential-impact-processors-power-family/
     -- Speculative execution?

IME/Pluton -- backdoors
    https://www.techrepublic.com/article/is-the-intel-management-engine-a-backdoor/
    https://www.techrepublic.com/article/why-the-nsa-may-not-need-backdoors/
    Disabled on new ThinkPads: https://www.theregister.com/2022/01/20/microsoft_amd_pluton_lenovo/

Heatbleed (2014) (occured in open source software)

Government
    https://rules.house.gov/bill/117/hr-4521

    https://www.theguardian.com/world/2014/feb/27/gchq-nsa-webcam-images-internet-yahoo
    https://archive.nytimes.com/www.nytimes.com/interactive/2013/09/05/us/documents-reveal-nsa-campaign-against-encryption.html
    https://www.propublica.org/article/the-nsas-secret-campaign-to-crack-undermine-internet-encryption
    https://www.nytimes.com/2013/09/06/us/nsa-foils-much-internet-encryption.html
    !! https://wikiless.org/wiki/Dual_EC_DRBG
    https://www.reuters.com/article/us-usa-security-rsa-idUSBRE9BJ1C220131220
    https://web.archive.org/web/20131223121638/http://blogs.rsa.com/news-media-2/rsa-response/
    https://www.technologyreview.com/2012/04/04/186902/how-china-blocks-the-tor-anonymity-network/
    https://www.nytimes.com/2016/09/03/technology/nso-group-how-spy-tech-firms-let-governments-see-everything-on-a-smartphone.html
    Leahy Law
    DeadHand and MonsterMind

Terror
    September 2001
    2001 Anthrax attacks

Privacy
    Apple and App Tracking Transparency
    https://www.flurry.com/blog/ios-14-5-opt-in-rate-att-restricted-app-tracking-transparency-worldwide-us-daily-latest-update/
    https://www.bloomberg.com/news/articles/2021-07-14/facebook-fb-advertisers-impacted-by-apple-aapl-privacy-ios-14-changes

Quantum computing
    https://sci-hub.se/10.1007/978-3-540-88702-7_1
    https://aapt.scitation.org/doi/abs/10.1119/1.1891170
    https://ieeexplore.ieee.org/abstract/document/8490169
    https://digitalcommons.dartmouth.edu/senior_theses/23/
    https://www.sciencedirect.com/science/article/abs/pii/S1361372317300519
    https://arxiv.org/abs/1804.00200

Surveillance
    https://www.nytimes.com/2022/02/10/us/politics/cia-data-privacy.html
    https://www.eff.org/deeplinks/2022/02/we-need-answers-about-cias-mass-surveillance

crowd supply boosts open hardware: linux magazine

{firewall}