17 July 2021 | New Scientist | 13A MICROSOFT-owned
tool powered by artificial
intelligence is designed to make
life easier for programmers,
but some developers say it may
be repurposing some of the
billions of lines of code it was
trained on without permission.
The tool, called CoPilot,
was released by GitHub, a
Microsoft subsidiary that is
used by millions of people to
share source code and organise
software projects. CoPilot
uses powerful neural network
tools developed by OpenAI to
solve programming problems
by scouring vast numbers
of examples of existing
solutions, both from GitHub
and elsewhere, and learning
how to create similar solutions.
It then suggests code based on
a human programmer’s work in
progress or English descriptions
of the functionality needed.
But sometimes CoPilot may
directly plagiarise its training
data, says Armin Ronacher
at software company Sentry.
He has found that it is possible
to prompt CoPilot to suggest
copyrighted code from the 1999
computer game Quake III Arena,
complete with comments fromthe original programmer.
“You can definitely make
it recite code that is almost
entirely in the training set
where there’s no originality
happening,” says Ronacher,
although he says it should be
possible to adjust CoPilot so that
it warns the user if code being
output is close to original work.A more difficult problem
to solve is that many of the
software projects that CoPilot
has been trained on are released
under free software licences
such as the General Public
License, or GPL, that only allow
derivative works if they are also
freely released. This doesn’t stop
them being used commercially.
“Free” in this context means
that people are free to modify
the code, but it does mean
that people using CoPilot would
in theory also be required to
release any source code they
create for other people to use.
However, the CoPilot website
says that copyright for the codeit generates belongs to the
programmer using it.
Some developers have already
taken action to protect their
work. Adrian Bowyer at RepRap,
an open source 3D printer
project, says CoPilot is a great
idea but anything it creates
must itself be open source.
He has altered the wording
of the licence under which
he releases software: “If any
part of RepRap covered by
the GPL is used to train any AI,
then all the products of that
AI must be released under
the GPL as free software.”
The CoPilot website says that
about 0.1 per cent of CoPilot
suggestions may contain
“some snippets” of verbatim
code from the training set.
Neil Brown at UK law firm
decoded.legal says that GitHub
has made a “bald assertion”
that it can analyse code in the
way CoPilot does under the fair
use copyright infringement
defence in the US, but the
position in the UK is less clear.
Yin Harn Lee at the University
of Bristol, UK, agrees that
CoPilot could be allowed under
US copyright law, which is less
prescriptive, but says it is a grey
area in the UK that needs to be
tested in court. The US fair use
is a broader, less defined and
more unpredictable defence.
UK copyright law tends to offer
a much clearer definition of
what is and isn’t acceptable.
“I’m very keen for it to
be tested, because I want
to know,” says Harn Lee.
Microsoft and OpenAI
didn’t respond to requests
for comment, while GitHub
declined to comment. ❚WHEN grown in tiny strands, ice can
bend and then return to its original
shape. These microfibres are the
most flexible form of ice ever made.
Most water ice is extremely rigid
and brittle, breaking easily rather
than bending. However, a single,
long crystal of ice can be far more
flexible. Limin Tong at Zhejiang
University in Hangzhou, China, and
his colleagues have used this quality
to fabricate the most elastic water
ice ever, close to the theoretical limit
of how flexible it can be.
They made their ice fibres using
water vapour piped into a small
chamber kept at a temperature
of -50°C. An electric field in the
chamber attracted water molecules
to a needle made of tungsten, where
they crystallised to build fibres up
to a few micrometres in diameter.
The researchers then cooled the
ice even further, to between -70°C
and -150°C, and measured the
elastic strain of the fibres, which
is a way of assessing how much
a material is being bent and
deformed. They found that these
fibres were more elastic than any
other water ice structures that have
been measured – some could nearly
be bent into circles, and all of them
returned to being straight lines
afterwards (Science, doi.org/gmxk).
Previously, the largest elastic
strain observed in ice was about
0.3 per cent, but this is about
10.9 per cent in the team’s ice fibres,
says Tong. The theoretical limit for
the elastic strain in water ice is
between 14 and 16.2 per cent.
When Tong and his colleagues
examined the ice strands in detail,
they found hints of the presence of
a second form of ice that is denser
than the type of ice making up the
majority of the fibres. The stress on
the bent part of the fibre may have
driven a transformation in the ice,
which means these fibres could
potentially help us understand
how such changes work. ❚Writing code could
be easier with
an AI assistantPhysics SoftwareLeah Crane Matthew SparkesMONSTA
R^ STUDIO/SHUTTE
RS
TOCK0.1%
of CoPilot suggestions may
reuse code from elsewhereAI programmer may be
reusing code without asking
New kind of ice
can curl and uncurl
without breaking