Zürcher Nachrichten - Does "vibe coding" make everyone a programmer?

EUR -
AED 4.35335
AFN 77.050797
ALL 96.614026
AMD 452.873985
ANG 2.121943
AOA 1087.00321
ARS 1723.800654
AUD 1.702936
AWG 2.136666
AZN 2.019869
BAM 1.955248
BBD 2.406031
BDT 145.978765
BGN 1.990709
BHD 0.449191
BIF 3539.115218
BMD 1.18539
BND 1.512879
BOB 8.254703
BRL 6.231008
BSD 1.194568
BTN 109.699013
BWP 15.630651
BYN 3.402439
BYR 23233.647084
BZD 2.402531
CAD 1.615035
CDF 2684.909135
CHF 0.915881
CLF 0.026011
CLP 1027.058063
CNY 8.240537
CNH 8.248946
COP 4354.94563
CRC 591.535401
CUC 1.18539
CUP 31.412839
CVE 110.234327
CZK 24.334287
DJF 212.720809
DKK 7.470097
DOP 74.383698
DZD 153.702477
EGP 55.903178
ERN 17.780852
ETB 185.572763
FJD 2.613371
FKP 0.863571
GBP 0.865754
GEL 3.194674
GGP 0.863571
GHS 12.974143
GIP 0.863571
GMD 86.533903
GNF 10372.164298
GTQ 9.16245
GYD 249.920458
HKD 9.257838
HNL 31.365884
HRK 7.536597
HTG 156.336498
HUF 381.328619
IDR 19883.141804
ILS 3.663335
IMP 0.863571
INR 108.679593
IQD 1553.453801
IRR 49934.560565
ISK 144.985527
JEP 0.863571
JMD 187.197911
JOD 0.840489
JPY 183.433247
KES 152.915746
KGS 103.662825
KHR 4768.236408
KMF 491.93733
KPW 1066.928941
KRW 1719.752641
KWD 0.36382
KYD 0.995519
KZT 600.800289
LAK 25485.888797
LBP 101410.128375
LKR 369.427204
LRD 219.593979
LSL 19.132649
LTL 3.500149
LVL 0.717031
LYD 7.495914
MAD 10.835985
MDL 20.092409
MGA 5260.173275
MKD 61.631889
MMK 2489.287708
MNT 4228.659246
MOP 9.606327
MRU 47.30937
MUR 53.852723
MVR 18.32658
MWK 2059.023112
MXN 20.70407
MYR 4.672854
MZN 75.580924
NAD 18.967522
NGN 1643.520192
NIO 43.508231
NOK 11.437875
NPR 175.519161
NZD 1.96876
OMR 0.458133
PAB 1.194573
PEN 3.994177
PGK 5.066955
PHP 69.837307
PKR 331.998194
PLN 4.215189
PYG 8001.773454
QAR 4.316051
RON 5.097064
RSD 117.111851
RUB 90.544129
RWF 1742.915022
SAR 4.446506
SBD 9.544303
SCR 17.200951
SDG 713.016537
SEK 10.580086
SGD 1.505332
SHP 0.88935
SLE 28.834661
SLL 24857.038036
SOS 677.454816
SRD 45.104693
STD 24535.182964
STN 24.493185
SVC 10.452048
SYP 13109.911225
SZL 19.132635
THB 37.411351
TJS 11.151397
TMT 4.148866
TND 3.37248
TOP 2.854135
TRY 51.47818
TTD 8.110743
TWD 37.456003
TZS 3052.380052
UAH 51.199753
UGX 4270.811618
USD 1.18539
UYU 46.357101
UZS 14603.874776
VES 410.075543
VND 30749.020682
VUV 141.680176
WST 3.213481
XAF 655.774526
XAG 0.014004
XAU 0.000244
XCD 3.203577
XCG 2.153028
XDR 0.815573
XOF 655.774526
XPF 119.331742
YER 282.508153
ZAR 19.136335
ZMK 10669.938133
ZMW 23.443477
ZWL 381.695147
  • SCS

    0.0200

    16.14

    +0.12%

  • RBGPF

    1.3800

    83.78

    +1.65%

  • CMSD

    -0.0400

    24.05

    -0.17%

  • BCC

    0.5100

    80.81

    +0.63%

  • JRI

    0.1400

    13.08

    +1.07%

  • CMSC

    0.0500

    23.76

    +0.21%

  • RIO

    -4.1000

    91.03

    -4.5%

  • RELX

    -0.3700

    35.8

    -1.03%

  • BCE

    0.3700

    25.86

    +1.43%

  • NGG

    0.2000

    85.27

    +0.23%

  • GSK

    0.9400

    51.6

    +1.82%

  • BTI

    0.4600

    60.68

    +0.76%

  • VOD

    -0.0600

    14.65

    -0.41%

  • RYCEF

    -0.4300

    16

    -2.69%

  • BP

    -0.1600

    37.88

    -0.42%

  • AZN

    0.1800

    92.77

    +0.19%

Does "vibe coding" make everyone a programmer?
Does "vibe coding" make everyone a programmer? / Photo: Lionel BONAVENTURE - AFP

Does "vibe coding" make everyone a programmer?

Can a complete tech novice create a website using everyday language on ChatGPT?

Text size:

That's the promise, misleading for some, of "vibe coding," the latest Silicon Valley catchphrase for an advance in generative AI that some say makes computer programming as simple as chatting online.

"You fully give in to the vibes, embrace exponentials, and forget that the code even exists," OpenAI co-founder and former Tesla employee Andrej Karpathy described in early February, in a message posted on X (formerly Twitter), using the term for the first time.

"I'm building a project or web app, but it's not really coding - I just see stuff, say stuff, run stuff, and copy paste stuff, and it mostly works," he said.

The developer and entrepreneur was referring to the new generative AI models that produce lines of code on demand in everyday language, through writing or speech.

The concept of "vibe coding" remained confined to the AI community until New York Times columnist Kevin Roose claimed to have created websites and apps without any knowledge of programming.

"Just having an idea, and a little patience, is usually enough," he wrote.

The ChatGPT and Claude interfaces can write an entire program line by line on demand, as can Gemini, which launched its dedicated version, Gemini Canvas, on Tuesday.

Other generative AI platforms specifically dedicated to coding have also made their mark in recent months, from Cursor to Loveable, or Bolt, Replit and Windsurf.

"Maybe, just maybe, we're looking at a fundamental shift in how software is created and who creates it," said online marketing specialist Mattheo Cellini on Substack.

"It's unlikely to make coding irrelevant, but it may change the way developers work," suggested Yangfeng Ji, professor of computer science at the University of Virginia.

"This could lead to some job displacement, particularly for those focused solely on basic coding tasks."

Even before "vibe coding," a downturn was being seen by some in IT employment as the first effects of generative AI began to be felt.

The sector shed nearly 10,000 jobs in the US in February, according to the Department of Labor, and its headcount is at a three-year low.

- Expertise needed? -

Among code novices, many find it hard to catch the vibe.

"People who do not have programming expertise often struggle to use these kinds of models because they don't have the right kinds of tools or knowledge to actually evaluate the output," said Nikola Banovic, professor of computer science at the University of Michigan.

On social media, the few newbies who report on their "vibe coding" quickly complain that it's not as easy as some want to believe.

Without mastering computing complexities like digital directories, runtime environments or application programming interfaces (APIs), it's hard to create an app that works.

Despite his coding knowhow, Claude Rubinson, a professor of sociology at the University of Houston-Downtown, wanted to create an application for his students two years ago without tinkering with the code generated by ChatGPT.

After a lot of trial and error, the app finally worked, but "I'm convinced it wouldn't have worked if I hadn't understood the code," which allowed him to guide the interface using the appropriate language.

This brought home the importance of the "prompt": mastering the request submitted to obtain the desired result.

"Programmers have certain levels of AI literacy that allows them to get what they want out of the models," said Banovic.

Everyday users "will not know how to prompt," h warned.

A.P.Huber--NZN